USING IMPROVEMENT SCIENCE TO IMPROVE INSTRUCTION DISTRICT-WIDE: A COMPARATIVE CASE STUDY By Kimberly Ann Jansen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Educational Policy- Doctor of Philosophy 2018 USING IMPROVEMENT SCIENCE TO IMPROVE INSTRUCTION DISTRICT-WIDE: A COMPARATIVE CASE STUDY ABSTRACT By Kimberly Ann Jansen A persistent problem in education has been around how to improve schools where the majority of students are not meeting academic expectations. Under No Child Left Behind (NCLB) implemented in 2001, schools where students failed to meet adequate yearly progress for five consecutive years were given the options to: (1) close and reopen as a charter school, (2) close and reopen under the direction of an outside organization, (3) close and reopen under the direction of state emergency management, (4) begin the following year with a new principal and a large proportion of new staff, or (5) begin the following year with another major restructuring plan that included professional development, new curriculum, reduced class sizes, and other related interventions. For the most part, these choices of strategies remained as the school turnaround strategies until 2015 with the passage of the Every Student Succeeds Act (ESSA). Overall these five strategies, prioritize interventions around the management of schools and fail to address student and teachers’ local contexts and individual knowledge. The passage of ESSA allowed for an increased role of states to shape school turnaround policy and design new programs and approaches. This dissertation examines one such program in Michigan, the Blueprint for Strategic Reconfiguration. The Blueprint for Strategic Reconfiguration calls on districts to develop district-level and building-level networks to develop and implement a coordinated vision of high-quality instruction. Unlike previous initiatives, the Blueprint draws on improvement science principles to work to improve instruction at scale across a district. Improvement science aims to answer the question “what works, for whom, and under what set of conditions?” (Bryk et al., 2015. p. 14). According to improvement science, change occurs when local change agents engage in multiple cycles of improvement design, testing, and implementation (Bryk et al., 2011). In this dissertation, I examine this improvement science-based approach to school turnaround and improving instruction at scale across the district. In particular, I answer the following research questions: (1) What is the intended theory of change of the Blueprint for Systemic Reconfiguration for influencing instruction at scale?; (2) How does each administrator’s orientation toward change influence the implementation of Blueprint for Systemic Reconfiguration in their district or school?; (3) How is the theory of change enacted in each district? How does this enactment reflect administrators’ orientations toward change?; and (4) To what degree and in what ways does the enactment of the theory of change influence teacher instructional practices and planning in each school? This embedded case study examines observational data from district leadership networks (n=2), school leadership networks (n=4), and teacher classrooms (n=44) as well as interview data from district superintendents (n=2), building principals (n=40, and teachers (n=44). It also drew on fall and spring teacher surveys regarding their reading instructional practices, collaboration opportunities, planning practices, and school environment. Results show: (1) the Blueprint is built on improvement science principles; (2) superintendents and principals have different orientation toward the change process, which typically fall into experimental science assumptions and improvement science assumptions; (3) the superintendent and principals’ orientation toward the change process influenced both the design and implementation of the Blueprint in their district and building; and (4) these different orientations toward change and the differing implementation strategies influenced teacher implementation of the district mandated literacy program. Copyright by KIMBERLY ANN JANSEN 2018 ACKNOWLEDGEMENTS I have been incredibly fortunate throughout my graduate studies to have the support of so many amazing people. Without the support, feedback, and encouragement from these people this work would not have been possible. First, I am extremely grateful for my committee members. Thank you to Dr. Kristy Cooper Stein for constantly supporting and challenging me over the past few years. I sincerely appreciate the force of calm and guidance that you have been as I progressed through the program. Additionally, I would like to thank Dr. Ken Frank for helping me grow as a scholar and researcher over the past three years. Through working on the SEMI project, conversations with you, and discussions at research group, I have learned how to think like a scholar. Thank you for being a model of thoughtful scholarship. Thank you to Dr. Madeline Mavrogordato for the support and critical feedback that consistently pushed my thinking and work in meaningful ways. And thank you to Dr. Corey Drake for joining my dissertation committee and providing thoughtful feedback. I truly appreciate the opportunity to work with you. Beyond my committee, I have been blessed over the past five years to be supported by many other faculty members at Michigan State and my graduate student colleagues. First, thank you to Dr. William Schmidt and his team for their constant support, advice, and willingness to provide feedback on my work. Dr. Schmidt thank you for providing me with great guidance beginning on my very first day on campus. I have truly appreciated the opportunity to learn through working with you. Additionally, Dr. Michael Sedlak thank you for the opportunity to study at Michigan State University and the extensive support that you have provided me during my time in the program. To my graduate student colleagues, thank you for exposing me to new v ideas and perspectives. It has been a joy to learn alongside all of you. More specifically, a big thank you to Jason Burns, Ben Creed, John Lane, Alyssa Morley, David Reid, and Rachel White for their constant support and feedback throughout the dissertation process. I would also like to thank the administrators at Hillside and Lakeside school districts as well as teachers at Cottonwood, Hickory, Rose, and Willow Elementary Schools who were so generous in sharing their time and experiences with me over the course of the 2016-2017 school year. Thank you to Dr. Grant Chandler for supporting this research and teaching me about the Blueprint. Additionally, I would like to thank Ashley Parrish and Maggie Keech for their assistance in transcribing the teacher interviews. Thank you to my friends and family who inspired the pseudonyms in this dissertation. To Kathy Burns, Lindsey Clark, Adrienne DeSutter, Vanessa Johnson, Anna Miller, Kristen Olin, and Ashley Parrish thank you for all of your love and encouragement from afar. Lastly, I would like to thank my family. Thank you to my parents and sister for their continuous support for all of my dreams. Your constant love has made my accomplishments possible. Finally, to my fiancé, Michael Evert, who was willing to move to a state that he had never visited before so that I could go back to school. Thank you for your endless support and love. Especially in the past few months as I finished this dissertation, thank you for keeping our house standing and the cats alive. I am forever grateful. I love you. I am forever indebted to these and so many other people, who I do not have the space to thank, for supporting my work in education and on this dissertation. Thank you for all of your guidance and encouragement. vi TABLE OF CONTENTS LIST OF TABLES .................................................................................................................... x LIST OF FIGURES ................................................................................................................. xi KEY TO ABBREVIATIONS .................................................................................................xii CHAPTER 1: INTRODUCTION ............................................................................................. 1 Research Questions ............................................................................................................... 6 Study Context: MI Excel Blueprint for Systemic Reconfiguration ..................................... 7 CHAPTER 2: REVIEW OF RELEVANT LITERATURE .................................................. 11 Literature Review ............................................................................................................... 11 History of School Turnaround Efforts................................................................................ 11 Instructional Core and the Problem of Scale ...................................................................... 18 Organizational Learning .................................................................................................... 19 Social Capital .................................................................................................................... 22 Social Networks ................................................................................................................ 23 Social Networks and Organizational Learning ................................................................... 25 Social Networks and School Reform ................................................................................. 26 Conceptual Framework ...................................................................................................... 28 Traditional Education Research Science – Experimental Science....................................... 29 Components of Improvement Science ............................................................................... 31 Summary ........................................................................................................................... 36 Improvement Science in Education ................................................................................... 37 The Usefulness of Improvement Science for Analyzing School Improvement Efforts ........ 43 CHAPTER 3: RESEARCH DESIGN .................................................................................... 45 Research Design and Research Questions .......................................................................... 45 Participants and Sampling Strategy ................................................................................... 47 Data Collection .................................................................................................................... 51 Direct Observations ........................................................................................................... 51 Documentation .................................................................................................................. 53 Interviews ......................................................................................................................... 54 Survey ............................................................................................................................... 56 Quantitative Measures ....................................................................................................... 57 Data Analysis ....................................................................................................................... 61 Research Question 1: Theory of Action for the Blueprint for Systemic Reconfiguration .... 61 Research Question 2: Administrator Orientation to Change and District Theory of Action 64 Research Question 3: Implementation of Each District’s Theory of Action ........................ 65 Research Question 4: Changes in Instructional Practice and Planning ................................ 66 Summary of Methods ........................................................................................................ 68 Limitations .......................................................................................................................... 70 Establishing Validity ........................................................................................................... 72 vii CHAPTER 4: THEORY OF ACTION AND LEADERSHIP ORIENTATIONS TO CHANGE ................................................................................................................................ 75 Theory of Action for the Blueprint for Systemic Reconfiguration .................................... 76 Theory of Action for the Instructional Infrastructure.......................................................... 82 Blueprint and Improvement Science .................................................................................. 88 Theories of Action in Practice............................................................................................. 89 Superintendent’s Role ....................................................................................................... 90 Ms. King – “We adhere to it with fidelity” ........................................................................ 92 Hillside Theory of Action for the Instructional Infrastructure ............................................ 97 Mr. Dixon – “Classrooms should be a petri dish” .............................................................. 99 Lakeside Theory of Action for the Instructional Infrastructure ......................................... 106 Implementation Orientations at the School Level ........................................................... 109 Ms. Davis – “Walk that line” ........................................................................................... 109 Ms. Nelson – “I understand it on a deeper level” ............................................................. 112 Ms. Wilson – “Doing it with purpose” ............................................................................. 115 Ms. Reed – “Not just our school but our district” ............................................................. 119 School Principal Summary .............................................................................................. 122 Chapter Summary ............................................................................................................. 123 CHAPTER 5: IMPLEMENTATION INTO SCHOOLS AND CLASSROOMS ............... 124 Mandated Literacy Instructional Policies ........................................................................ 124 Fidelity Program ............................................................................................................. 125 Professional Program ...................................................................................................... 129 Enactment Processes ......................................................................................................... 135 Resources ........................................................................................................................ 136 Training .......................................................................................................................... 138 PSP Meetings .................................................................................................................. 143 Monitoring ...................................................................................................................... 144 Monitoring with Flexibility ............................................................................................. 155 Summary ......................................................................................................................... 156 Teacher Implementation of District Literacy Expectations ............................................ 162 Hillside Implementation .................................................................................................. 162 Lakeside Implementation ................................................................................................ 171 Role of Knowledge in Implementing Tendencies............................................................. 180 Quantitative Examination of Implementation................................................................... 182 Summary ........................................................................................................................... 185 CHAPTER 6: DISCUSSION, IMPLICATIONS, AND CONCLUSION ............................ 188 Summary of Findings ........................................................................................................ 188 Blueprint for Systemic Reconfiguration Theory of Action ............................................... 188 Leadership Assumptions of Change and Implementation ................................................. 189 Influencing Teaching Practices at Scale ........................................................................... 190 School Turnaround and Improvement Science ............................................................... 192 What does good teaching at scale mean? ......................................................................... 196 Implications ....................................................................................................................... 201 Implications for Policymakers ......................................................................................... 201 viii Implications for School Practitioners ............................................................................... 202 Implications for Researchers ........................................................................................... 205 Future Research ................................................................................................................ 206 Conclusions ........................................................................................................................ 207 APPENDICES ....................................................................................................................... 209 APPENDIX A: Timeline of Blueprint Installation .......................................................... 210 APPENDIX B: Observation Protocol ............................................................................... 211 APPENDIX C: Sample Vision of High Quality Reading Instruction ............................. 212 APPENDIX D: Fall Superintendent Interview Protocol ................................................. 215 APPENDIX E: Spring Superintendent Interview Protocol ............................................. 217 APPENDIX F: Fall Building Principal Interview Protocol ............................................. 218 APPENDIX G: Spring Building Principal Interview Protocol ....................................... 220 APPENDIX H: Teacher Interview Protocol – Round 1 .................................................. 222 APPENDIX I: Teacher Interview Protocol – Round 2 .................................................... 223 APPENDIX J: Teacher Interview Protocol – Round 3.................................................... 224 APPENDIX K: Teacher Interview Protocol – Round 4 .................................................. 225 APPENDIX L: Fall Teacher Survey................................................................................. 226 APPENDIX M: Spring Teacher Survey ........................................................................... 236 APPENDIX N: Theory of Change for the Blueprint for Systemic Reconfiguration ...... 247 APPENDIX O: Sample Instructional Infrastructure Planning Tool – Floor 1 .............. 248 APPENDIX P: Lakeside School District HQI Walkthrough Observation Form ........... 249 APPENDIX Q: Hillside Curriculum Framework- Unit 2 ............................................... 250 APPENDIX R: Lakeside School District High Quality Instruction Implementation Guide ........................................................................................................................................... 260 APPENDIX S: CAFE Menu Bulletin Board (from The Daily Cafe, 2018) ..................... 261 APPENDIX T: CAFE Link in Lakeside Implementation Guide .................................... 262 APPENDIX U: Cottonwood May PSP Form ................................................................... 263 APPENDIX V: Sample Mandated Anchor Chart ........................................................... 264 REFERENCES ..................................................................................................................... 265 ix LIST OF TABLES Table 3.1: Participant Information ............................................................................................. 50 Table 3.2: Hillside School District Measure of Aligned Teaching ............................................. 58 Table 3.3: Lakeside School District Measure of Aligned Teaching ............................................ 58 Table 3.4: Measure of Planning with District Resources ............................................................ 60 Table 3.5: Measure of Perceived Coherence .............................................................................. 60 Table 3.6: Measure of Perceived Professional Fit ...................................................................... 61 Table 3.7: Summary of Methods ............................................................................................... 68 Table 4.1: Summary of Administrator Interview Excerpts that Aligned with Either Experimental Science or Improvement Science Viewpoints ............................................................................ 91 Table 5.1: Experimental Science and Improvement Science Orientation of District Literacy Program Implementation Strategies (as defined by Lewis, 2015, p. 56) ................................... 157 Table 5.2: Advanced Literacy Instructional Knowledge and Teacher Implementation Practices ............................................................................................................................................... 180 Table 5.3: Spring Aligned Teaching Descriptive Statistics ...................................................... 183 Table 5.4: Summary of Linear Regression Results for Spring Aligned Reading Instruction ..... 184 x LIST OF FIGURES Figure 1.1: Graphic Organizer for MI Excel’s Blueprint for Systemic Reconfiguration ............... 9 Figure 2.1: Scale-up of Knowledge: Contrast of Paradigms (from Lewis, 2015, p. 56) .............. 36 Figure 2.2: Lesson Study Process (from Lewis et al., 2006, p. 4) ............................................... 42 Figure 3.1: Design of Case Study (adapted from Yin, 2014, p. 50) ............................................ 47 Figure 3.2: Blueprint for Systemic Reconfiguration Theory of Action (Chandler & Frank, 2017) ................................................................................................................................................. 63 Figure 4.1: Theory of Action for Instructional Infrastructure ..................................................... 83 Figure 4.2: Theory of Action for Staff Development in Instructional Infrastructure ................... 87 Figure 4.3: Hillside School District Theory of Action for Influencing Reading Instruction ........ 97 Figure 4.4: Lakeside School District Theory of Action for Influencing Reading Instruction .... 106 Figure 5.1: Hickory Elementary Literacy Advice Network ...................................................... 164 Figure 5.2: Willow Elementary Literacy Advice Network ....................................................... 168 Figure 5.3: Cottonwood Elementary Literacy Advice Network ............................................... 173 Figure 5.4: Rose Elementary Literacy Advice Network ........................................................... 179 xi KEY TO ABBREVIATIONS DN District Network BN Building Network PSP Problem-Solving Protocol PMT Performance Management HQI High-Quality Instruction, usually in reference to HQI walkthroughs SSN Student Support Network SST Student Support Team xii CHAPTER 1: INTRODUCTION For the past few decades, schools have experienced a vortex of change caused by ambitious yet disconnected reforms that call on districts to constantly alter their practices. These reforms tend to affect superficial processes, such as course schedules and school hours. However, large changes in the core of what teachers and students actually do together in classrooms rarely affect more than a select few of classrooms in any district. Even when these reforms do affect change, the changes seldom last very long (Elmore, 2004). Within the current vortex of change, basic conceptions of knowledge, of teachers’ and students’ roles in constructing knowledge, and of the role of classroom-level and school-level structures in enabling student learning remain relatively static (Elmore, 2004). What has resulted from all of these disconnected reform efforts in the past few decades is that schools are constantly changing without achieving any fundamental change (Elmore, 1996). Lasting school improvement will occur only if innovations can improve the instructional core of schooling, which describes the interactions between teachers, students, and learning materials (Cohen & Ball, 1999; Elmore, 1996). Throughout the past few decades, policymakers and local administrators have enacted well-intentioned policy with the hopes of improving school performance, yet these policies rarely if ever reach the instructional core of what teachers and students do together in the classroom (Elmore, 2008). For example, in the 1990s and 2000s under the Goals 2000 and No Child Left Behind (NCLB) policies, states created and adopted curriculum standards, which were intended to change the content being taught in each classroom and hold schools accountable for the level of the content (Neal & Schanzenbach, 2010). However, the effect of standards to improve school performance is dependent on having appropriate materials in the classroom, 1 adequate teacher knowledge of the content and standard requirements, and student willingness to engage with the content prescribed in the standards (Elmore, 2008). In fact, Mayrowetz (2009) finds in a study of 12 mathematics teacher that half of them did not state that the standards influenced their instruction in any way. Additionally, numerous other studies find that NCLB led to counterproductive gaming responses by teachers, such as narrowing the curriculum to basic skills and reclassifying students into special education (i.e. Figlio & Getzler, 2002; Neal & Schanzenbach, 2010). After curriculum standards failed to improve school performance, in December 2009, under the federal guidelines included from the Race to the Top (RTTT) fund, standards-based teacher evaluation became a focal point for policymakers in attempting to influence the quality of classroom instruction (Donaldson & Papay, 2015). However, yet again the effectiveness of teacher evaluation to influence the instructional core is dependent on the quality of the evaluation practices and school leaders to affect teachers’ knowledge and skill (Elmore, 2008). In fact, research has shown that teacher evaluation alone is a weak lever for significant improvements in teaching and student learning (Smylie, 2014). Research from Cincinnati and Chicago finds that evaluation systems must be paired with professional development and coaching from principals in order to improve student outcomes (Steinberg & Sartain, 2015; Taylor & Tyler, 2015). As demonstrated from the examples above, when focusing only on one component of the instructional core – either the curriculum or the teacher- only minimal change is probable in classroom operations. Effective policy must attend to all three components of the instructional core in order to bring about instructional improvement (Elmore, 2008). Over the past two decades, perhaps the most drastic example of school improvement has been around school turnaround. School turnaround reform involves dramatic, multi-dimensional 2 and comprehensive change in school operations with the goal of improving the academic performance of the same student population in a compressed time period (Duke, 2012; Hassel, Hassel, Arkin, Kowal, & Steiner, 2006; Kutash, Nico, Gorin, Rahmatullah, & Tallant, 2010). Turnaround requires a change in both the direction and control of the school (Hassel et al., 2006). This drastic change often requires the replacement of building leaders and staff, intensive professional development, and changes in the school’s operations. While the replacement of staff will result in new teachers in the classroom and new leaders in the school, these changes do not necessarily influence the actual practices of how teachers interact with the curriculum and students across the school building in any coordinated way. Like the standards movement and teacher evaluation policies, school turnaround policy focuses only on one component of the instructional core. School turnaround policy centers on improving the skill of the teacher in the classroom but ignores the role of the curriculum and student engagement with the content in determining instructional effectiveness. This omission has stunted the effectiveness of school turnaround policy to bring about improved student learning at scale. Cohen and Ball (1999) summarize how policy, like school turnaround, fails to influence teacher instructional practice: The environments in which practice and policy intersect lack many of the resources that would support fruitful interaction, including a shared language of professional discourse, traditions of common work on teaching and learning, opportunities for professional learning, social and economic support for demanding instruction, and professional norms and incentives that support improvement. Instead these environments are marked by rapidly changing policy agendas, diffuse and often divergent guidance for instruction, 3 deep disagreement about the ends and means of schooling, and inconsistent support for instructional improvement (p. 14) In order to affect instructional practice at scale across a school district, policy will need to provide resources to allow for social interactions among building practitioners and district policymakers in order to develop these defined necessary work environments for promoting change at scale (Finnegan & Daly, 2012). Complicating school turnaround efforts is the fact that school districts are not simple organizations. School districts should be understood as systems with multiple units and levels, with central offices, schools, departments, and classrooms. School districts also have multiple dimensions with intersecting structures, social relationships, politics, and cultures. All of these units, levels, and dimensions are interactive and mutually influential parts of the whole system (Smylie, 2016). For this reason, focusing on developing change throughout the district and not a singular school is integral to supporting instructional improvement since school operations are dependent on district contexts. This research suggests that in order to influence change in the instructional core, policy must consider the entirety of system components, relationships among these components and organizational members, and the interactions between teachers, students, and the curriculum within classrooms throughout the organization. With the passage of the Every Student Succeeds Act (ESSA) in December 2015, now is time to focus on building knowledge of how to effectively turn around entire districts rather than individual schools (Wong, 2016). Under ESSA states have renewed flexibility in designing educational policy. While ESSA does require states to continue to identify and intervene in the bottom five percent of schools, the law grants states considerable discretion in deciding what action to take and developing any interventions (Klein, 4 2015). In addition, ESSA provides considerable financial resources for states to dedicate toward school turnaround efforts. The law permits states to set aside up to seven percent of their Title I funds for school turnaround efforts, which is an increase from the four percent permitted under the previous law NCLB (Klein, 2015). The flexibility afforded in ESSA provides a unique opportunity for states to experiment with locally developed interventions for school turnaround. This dissertation studies one such program, the Blueprint for Systemic Reconfiguration, which uses improvement science principles focused on leading improvements in instructional practices at scale throughout the district. Recently, scholars have developed, improvement science, a promising structure for creating and implementing innovations at scale across several organizations. Improvement science refers to a methodology for inquiries to improve practice (Bryk et al., 2015, p. 10). It is based on cycles of improvement, in which researchers and local actors jointly design, test, and implement interventions (Bryk et al., 2011). Researchers at the Carnegie Foundation are currently examining the concept of networked improvement communities (NICs) as a structure to promote responsive and meaningful change in organizations. NICs are intentionally organized groups (or networks) of diverse stakeholders from many different organizations that meet regularly to define common organizational problems, design possible interventions, and test out any designed interventions in members’ local contexts (Bryk, Gomez, Grunow, & LeMatheiu, 2015). The NIC structure may have promise in attempting to promote change at scale not just across several organizations, but also throughout an entire district and its sub-organizations. The Blueprint for Systemic Reconfiguration, developed by the MI Excel Statewide System of Support Program in the Michigan Department of Education (MDE), provides an alternative structure to school turnaround. The Blueprint for Systemic Reconfiguration uses 5 nested NICs at the district and school levels to attempt to promote change in the instructional core across the entire district in ways aligned with the district-developed vision of high-quality instruction (Frank & Chandler, 2016). The purpose of this dissertation is to examine how an alternative framework for school turnaround – in this case, MI Excel’s Blueprint for Systemic Reconfiguration – draws on improvement science principles to influence instructional practice at scale throughout low-performing districts. More specifically, this study examines how administrator orientations toward change influenced the design and implementation of Blueprint components in districts, schools, and classrooms. This dissertation also studies the extent to which teachers altered their instructional practice in alignment with the district’s vision of high-quality reading instruction and what influenced them to make these changes. Collectively, these points of inquiry examine how the Blueprint, an improvement science-based innovation focused on the instructional core diffused throughout the systems in these school district. This dissertation provides insight into how administrator orientations toward change and district systems can influence instructional change throughout a complex organization. Research Questions The overarching research question for this study is: How, if at all, does an improvement science-based reform influence the instructional core at scale throughout a school district? I answer this question through four sub-questions drawing on embedded case studies of four elementary schools in two districts participating in MI Excel’s Blueprint for Systemic Reconfiguration initiative: 1. What is the intended theory of change of the Blueprint for Systemic Reconfiguration for influencing instruction at scale? 6 2. How does each administrator’s orientation toward change influence the implementation of Blueprint for Systemic Reconfiguration in their district or school? 3. How is the theory of change enacted in each district? How does this enactment reflect administrators’ orientations toward change? 4. To what degree and in what ways does the enactment of the theory of change influence teacher instructional practices and planning in each school? Study Context: MI Excel Blueprint for Systemic Reconfiguration Designed and organized by the MI Excel Statewide System of Support (a division of the Michigan Department of Education), the Blueprint for Systemic Reconfiguration was a systems- focused program that focused on both district and school level policies and practices. Frank and Chandler (2016), two of the architects of the initiative, describe the purpose of the Blueprint by noting “The Blueprint was about designing and strengthening school systems as learning organizations for turnaround” (p. 7). A core belief behind the Blueprint was that turnaround requires instructional improvement at scale (Chandler, 2016). Consequently, the design of the Blueprint focused district and school building work on designing the systems necessary to support organizational learning around whichever innovation the district chooses to employ. At the district level, the work of the Blueprint called for the creation and installation of organizational structures and routines that work to improve instructional practice at scale throughout the district. At the building level, the Blueprint called for instructional leaders (i.e. principals, coaches, and specialists) to design instructional routines as well as procedures that make commonplace interactions between these instructional leaders and teachers in order to support the implementation of these instructional routines at scale throughout the school building (Frank & Chandler, 2016). 7 As illustrated in the program’s theory of action shown in Figure 1.1, the Blueprint utilized four district-level systems, three building-level systems, and three intermediary driver systems in order to implement practices that support the improvement of instructional practices at scale. These systems were created, implemented, and maintained through guidance from the nested networks at the district-level and school-levels. The first network formed when launching the Blueprint in a school district is the District Network (DN). MI Excel recommended that the superintendent, central office administrators, building administrators, selected teachers, selected board members, and union representatives serve on the DN. The DN worked on establishing four systems at the district level: talent management, instructional infrastructure, leadership network, and student support services. The talent management system focused on finding and retaining the most effective staff. The instructional infrastructure system coordinated the district’s instructional resources, learning and supports through creating subject-specific visions of high quality instructional practices, instructional materials and assessments, and an instructional improvement network. The third system, the leadership network, established formal structures that re-distribute responsibility for improving instruction to all levels of the organization. More specifically the leadership network defined a partnership between central office and building leaders, redesigned positons in the central office, and coordinates structure and policy shifts. Lastly, the student support network focused on implementing and supporting student support systems that meet social, emotional, health, and nutritional needs of students (Frank & Chandler, 2016). 8 Figure 1.1: Graphic Organizer for MI Excel’s Blueprint for Systemic Reconfiguration Once the DN established its four systems, each school building formed its own Building Network (BN), as shown at the bottom of Figure 1.1. Each BN organized three building-level routines: leveraging district systems, instructional leadership, and teacher collaboration. The leveraging district systems routine focused on ways that the building principal utilized district supports such as coaches and professional development. The instructional leadership routine supported teacher learning and the implementation of high-quality instructional practices. Finally, the teacher collaboration routines focused on developing increased opportunities for teachers to interact and collaborate with each other and coaches around their instruction. 9 The drivers worked in between the two levels of nested networks to connect their work. The first driver is the problem-solving (PSP) driver. This driver addressed the process for facilitating data conversations throughout the district. This process established a norm of inquiry, the analysis of multiple measures of data and the development of an action plan. The second driver was the communication driver. This system defined the process in which Blueprint content was communicated across members of the district. The last driver, the performance management driver, established the norms for measuring both the implementation of the Blueprint as well as the performance of each school in comparison with it goals (Frank & Chandler, 2016). Given the Blueprint’s strong alignment with social capital and organizational learning theories along with its foundation in improvement science, this turnaround program provided an innovative, research-based alternative to previous policies shaping school improvement efforts. Each component of the Blueprint and the theory behind it will be further explored below. 10 CHAPTER 2: REVIEW OF RELEVANT LITERATURE Literature Review History of School Turnaround Efforts Prior to the development of the Blueprint for Systemic Reconfiguration, school turnaround had already played a prominent role in educational policy for the past two decades. School turnaround is the dramatic improvement of student and teacher performance in a short amount of time (Frank & Chandler, 2016). In 2001, the No Child Left Behind (NCLB) legislation brought school turnaround reforms to the federal level (Duke, 2012). NCLB tied federal funds to accountability measures that required low-performing schools to improve their student performance on standardized tests under increasingly punitive sanctions (Duke, 2012). These accountability measures required schools to have a minimum percentage of students meet grade-level state standards known as adequate yearly progress (AYP). Schools that failed to meet AYP requirements for five consecutive years entered the school restructuring process designed to promote school turnaround (Duke, 2012). Under restructuring, during the fifth year, district leaders developed a school-restructuring plan for the school. If after the sixth year, the school still failed to make AYP, then the district would implement the plan (Hassel et al., 2006). NCLB allowed districts to choose from one of five restructuring options (Hassel et al., 2006): (1) chartering, by which the district closes the school and contracts out the operations of the school to a charter operator; (2) contracting, through which the district closes the school and the school reopens under the direction of an outside organization, such as an emergency management organization; (3) state takeover, in which the district turns over operations of the school to the state; (4) new leader and staff, where the district retains control over the school, but the district replaces the principal and a large proportion of the staff; and lastly (5), other major restructuring, 11 which includes plans for professional development, new curricula, pedagogical changes, reduced class and school size, team teaching, or a combination of these interventions. These restructuring options would later provide the framework for the next administration’s policies on school turnaround (Duke, 2012). The Obama administration continued this federal focus on school turnaround. In 2009, the American Recovery and Reinvestment Act (ARRA) act reserved $4.35 billion for the Race to the Top (RTTT) fund, which created a competitive grant program for states to vie for federal funding (U.S. Department of Education, 2009). One of the four focus areas of the competition was turning around persistently low-performing schools (U.S. Department of Education, 2009). Along with the RTTT funds, ARRA allocated an additional $3.55 billion to states through the School Improvement Grants (SIG) program in 2009 (Kutash et al., 2010). States were granted the power to decide how much money each school would receive based on district applications. Finally, also funded by ARRA, an additional $0.65 billion in competitive grants were distributed through the Investing in Innovation (i3) Fund. This program awarded funds to both nonprofit organizations and school districts in order to expand evidence-based approaches to improve student achievement (Kutash et al., 2010). These funding sources provided the federal government with some power to promote school turnaround efforts at the state and district levels (Kutash, 2010). In December 2009, the US Department of Education (DOE) released specific guidelines for school turnaround policy replacing the previous NCLB guidance (Duke, 2012). The report defined the target for school turnaround reforms as “persistently low-achieving schools” outlined as schools that ranked in the bottom 5% of a state’s schools in student academic achievement or high schools that graduated less than 60% of their students (US Department of Education, 2009). 12 The DOE also established four options for school turnaround (including a model also known as school turnaround) (Duke, 2012): 1. School closure – the persistently low-achieving school closes and students enroll in a nearby school. 2. Restart Model – the persistently low-achieving school is closed and operations are turned over to a charter operator or emergency management organization. 3. Transformation Model – the district replaces the principal in the persistently low- achieving school and improves school operations through curriculum reform, professional development, extended learning time, and other reforms. 4. Turnaround Model – the district replaces the principal in the persistently low-achieving school, rehires no more than half of the existing staff, chooses a new governance structure, and implements reforms similar to those in the transformation model. In the first round of SIG grants during the 2010-2011 school year, which utilized these DOE guidelines, three-quarters of the schools implemented the transformation model, while 20% chose the turnaround model, 4% chose the restart model, and 2% of schools were closed (Anrig, 2015; Dee, 2012). The architects of these school turnaround interventions assume that poor school performance is caused by a combination of: (1) poor linkages between state standards and various components of education, such as curriculum; (2) failure of state education authorities and local education agencies to take action on behalf of students in low-performing schools; (3) lack of capacity in the leaders of the schools; (4) lack of capable staff to support student achievement; and (5) previous school improvement efforts that relied on “piecemeal” reforms that neglected the larger issues (Duke, 2012; Rogers-Chapman, 2013). Consequently, for 13 dramatic improvements in student achievement to occur, turnaround policy suggests that interventions must target these problems. Outcomes of school turnaround efforts. Overall the outcomes of the past 15 years of turnaround efforts have resulted in mixed experiences for both the students and the professionals in low-performing schools. In a review of the literature, Duke (2012) finds that overall turnaround reforms have altered the learning conditions for students in these schools through increased use of interim tests to monitor student progress, increased instructional time in mathematics and reading, and extended school days to allow for more instructional time. Post turnaround, interviews with teachers from multiple studies report that the reform improved the school environment with decreased student fights and drug use along with improved student attendance and discipline (Cucchiara, Rooney, & Robertson-Kraft, 2015; Hamilton et al., 2014). In terms of student achievement changes, results have been mixed and erratic (Malen et al., 2002). When studying schools in California that were awarded SIG grants during the 2010- 2011 school year, Dee (2012) found some overall improvement in student achievement with most of the achievement growth concentrated in schools that adopted the turnaround model. Anrig (2015) found that schools that were awarded a SIG grant tended to experience gains in student achievement and decreases in the percentage of students in the bottom proficiency category, as compared with similar schools that did not receive the grant. However, the gains were modest and difficult to sustain. He found no observable difference in results between schools that implemented the transformation and turnaround reforms (Anrig, 2015). In Michigan in a case study completed 3 years after the initial cohort of schools received the SIG grants in the fall of 2010, Rice and colleagues (2015) found no significant difference in student achievement between SIG schools and comparison schools that did not receive the grants. 14 For teachers, other studies found that turnaround reform resulted in substantial human costs. In a case study of six schools, Rice and Malen (2003) found that the replacement of over half of the teachers created a brain drain, as many of the higher quality and more experienced teachers left their schools. Researchers have found that many teachers’ departures were voluntarily, as they felt insulted by the reform and/or afraid of the instability of the school’s future (Hamilton et al., 2014; Rice & Malen, 2003). Moreover, studies have shown that newly hired teachers often lacked experience and the training and skills in core academic subjects like mathematics and science (Dee, 2012; Hamilton et al., 2014; Rice & Malen, 2003). In a 2-year study of a district with six schools undergoing school turnaround, NCLB’s fourth restructuring model, in which the district replaces the school’s principal and at least half of its teachers, Rice and Malen (2003) found that nearly 75% of the newly hired teachers were entering their first year of teaching. Additionally, many of them had not yet attained certification and consequently were required to enroll in after-school and weekend courses. Others similarly have found that many turnaround schools also were forced to hire alternatively certified novice teachers (Hamilton et al., 2014). Moreover, in years 2 and 3 of the reform, teachers continued to elect to leave the school at higher rates than prior to turnaround (Hamilton et al., 2014). Rice and Malen (2003) describe additional human costs in school turnaround efforts based on interviews with 106 teachers. They find that the turnaround reform resulted in substantial task costs as teachers were required to commit increased time and effort because the influx of new teachers created a void of local knowledge and all of the pre-existing routines and procedures had to be reestablished. Additionally, teachers reported social costs of working in these environments because of the dismantling of their previous professional networks and struggling to reestablish these relationships with the new staff in new professional environments 15 that were characterized by suspicion and distrust. They summarized the teachers’ experiences as “simply put, reconstituted staffs had more intense work demands and fewer professional reserves” (p. 652). Finally, the reconstitution resulted in psychological costs for the teachers who worked in this setting. Teachers mourned the loss of their colleagues who left and felt lower levels of professional efficacy and self-worth (Rice & Malen, 2003). Similarly, through interviews and focus groups with 86 teachers in 13 schools, Cucchiara and colleagues (2015) found substantial variation in how teachers experienced working life in turnaround schools. In the schools where teachers have had positive experiences with the reform (4 schools), organizational leaders established clear organizational procedures, supported teacher efforts, listened to teacher input on school-level decisions, and held teachers and students accountable to clearly defined expectations. In schools where teachers found the turnaround reform to be challenging (5 schools), teachers felt that organizational leaders monitored them for compliance to arbitrary rules and overly scripted curriculum, changed plans repeatedly, and generally worked in opposition to their efforts. In these challenging settings, teachers felt powerless to put to use their professional knowledge (Cucchiara et al., 2015). The Blueprint for Systemic Reconfiguration promotes a different option for school turnaround. As described above, the Blueprint specifically focuses on creating district structures to influence instructional practices and student support systems at scale. Unlike the previous options, the Blueprint draws on the expertise that current district and school leaders and teachers hold regarding the district, its students, and its community. By organizing these stakeholders into networks developed to plan and implement school improvement efforts, the Blueprint provides a structure that capitalizes on this context-specific human and social capital already existing in the district and focuses it on improving instruction (Frank & Chandler, 2016). 16 Under the Blueprint teachers and staff hold leadership roles that influence the district’s turnaround efforts. This approach should not result in the same human costs found by Rice and Malen (2003) as a result of the previous turnaround policies. Instead, the Blueprint has the potential to empower rather than demoralize a district’s staff. Additionally, as discussed in more depth below, the Blueprint deviates from previous models of turnaround by drawing on improvement science principles to promote a systems approach that includes the district in school turnaround efforts rather than isolating low-performing schools. Potential of District-Focus. In the previously discussed school turnaround reform policies, the focus had been placed on improving the school building and ignored the role of the district in possible support of turnaround efforts. In fact, the district was often seen as a cumbersome obstacle to school turnaround (Trujillo, 2016). With the shift from No Child Left Behind’s strict sanctions to the flexibility in the Every Student Succeeds Act (ESSA), there is an opportunity for a more systemic focus in effecting change in efforts to improve student achievement in low-performing schools (Finnigan & Daly, 2016). Long-lasting change is most likely to occur when districts design multifaceted, holistic levers of change that fit into a district’s context and capabilities and address their unique organizational problems (Smylie, 2016). One problem in many low-performing districts is professional isolation. These districts lack the structures that allow for teachers and school leaders to communicate across grade levels and subject areas about student progress (Wong, 2016). Having a systems-focus requires district leaders to direct their attention to not just one part of the organization, but to the entire system as well as the relationships between the organization and its environment (Smylie, 2016). It also requires attention to all of the stakeholders in the organization as well as to the connections between them. This requires an 17 investigation of both the horizontal and vertical coherence within the system (Daly & Finnigan, 2016). As summarized by Daly and Finnegan (2016), “The work of district reform is at its core a social act that is grounded in sense-making and co-construction among a wide set of actors” (p. 239). In its design, the Blueprint uses two layers of networks to facilitate collaborative sense- making around the district’s current struggles and planned interventions. By developing a district-wide common understanding of problems and planned solutions, building and district- level staff will be able to share knowledge and resources throughout the turnaround process. Instructional Core and the Problem of Scale Cohen and Ball (1999) define the instructional core as interactions between students, teachers, and instructional materials. In order for substantial learning to occur, the interactions among all three elements must function well. A teacher’s background, knowledge, and beliefs influence how she understands and responds to both instructional materials and students. Likewise, the students’ previous experiences, knowledge, and engagement are also important for meaningful instruction. Additionally, interactions between students shape the resources available for learning inside of a classroom. Lastly, the chosen materials and curriculum influence instruction. How teachers and students interact with the learning materials represents the enacted curriculum inside of a classroom (Cohen & Ball, 1999). Elmore (2004) argues that school improvement efforts rarely touch the instructional core of schooling. Most school improvement interventions target only one part of the instructional core, the teacher, which is logical since teachers’ knowledge, skills, and beliefs hold a distinctly important role in interpreting the material and interactions with students. As Elmore (2008) summarizes “We tend to focus more on what the teacher is doing in front of the classroom than we do on the work that is actually on top of the student’s desks” (p. 3). However, if interventions 18 do not change the way that teachers interact with the other two components of the core- for example how they plan class tasks, see student work, and/or how they interpret students’ ideas - then they are unlikely to truly impact teaching practice or student learning (Cohen & Ball, 1999). The fundamental problem with school improvement work is that any attempts to change the existing patterns in the instructional core of education are usually unsuccessful on anything more than a small scale (Elmore, 1996). Elmore defines the problem of scale as the principle that “innovations that require large changes in the core of educational practice seldom penetrate more than a small fraction of US schools and classrooms, and seldom last for very long when they do” (p. 1-2). The task of future school improvement policy will be to not broaden the work of teaching with new initiatives that introduce more incoherence into the system. Instead any new school improvement policy must deepen the work of teaching by promoting a greater focus and coherent culture around instructional practice (Elmore, 2008). Organizational Learning Theory holds that in order to change instructional practice at scale, organizational learning must occur around the vision of high quality instruction. Organizational learning is undertaken by members in order to achieve organizational purposes and renew organizational operations (Boreham & Morgan, 2004). Organizational learning can be understood as the processes necessary to make the resources exchanged through relationships actionable for the benefit of the organization (Mohrman et al., 2003). It is important to distinguish organizational learning from individual learning. Organizational learning is not simply the process of aggregating the knowledge of all of the individuals in an organization but has to involve a process of synthesizing pre-existing organizational knowledge with individual learning to create a new collective understanding (Leithwood, Leonard, & Sharratt, 1998; Nicolini & Meznar, 19 1995; Resnick & Spillane, 2006). It involves the shared construction of knowledge that becomes common to all members of the school organization (Louis, 2007). An organization’s ability to detect and correct problems including counterproductive norms and practices is a key feature of organizational learning (Finnegan & Daly, 2012). Successful organizational learning will result in changes in organizational norms and members’ behaviors (Finnegan & Daly, 2012). Organizational learning stresses the notion of learning as a process rather than learning as the product of acquiring chunks of information (Hager, 2005). Schechter and Qadach (2012) argue that organizational learning consists of four processes: information acquisition, information distribution, information interpretation, and organizational memory. During the first phase, information acquisition, organizational members obtain knowledge from their social ties (Schechter & Qadach, 2012). The second process, information distribution, involves the sharing of information that leads to common understanding among organizational units and members. This sharing can be done through memos, letters, informal conversations, and/or reports (Schechter & Qadach, 2012). More specifically, in schools, information can be distributed in standards, professional development materials, and accountability protocols (Resnick & Spillane, 2006). Third, information interpretation is a socio-cognitive process that involves the exchange of views and attitudes to create a common, shared understanding of the distributed information (Crossan, Lane, & White, 1999; Schechter & Qadach, 2012). In this process, the meanings and perspectives of individual members are combined to create new collective understandings (Herrenkohl, 2008). Finally, the process of organizational memory involves storing organizational experiences and knowledge into the collective for future use, such as written into policies and standard operating procedures, in informal conversation between organization members, and new expectations for practice (Knapp, 2008; Schechter & Qadach, 2012). Once 20 the learning is encoded into the organizational memory, it becomes a collective resource for all members of the organization and guides future actions (Cross et al., 1999; Boreham & Morgan, 2004). The foundation of these processes is the social environment of the organization and the relational practices and dialogues between organizational members (Boreham & Morgan, 2004). In traditional school organizations, organizational learning processes are rare. As Louis (2007) summarizes, “teachers bring individually held knowledge from their prior experiences and training that is often difficult for colleagues to access and utilize. Isolating teachers in individual classrooms, departments, and grade level configurations mitigates against the effective sharing of individually held knowledge” (p. 6). In order for organizational learning to happen in schools, teachers and administrators must work collaboratively. In analyses of multiple case studies of schools, scholars have found that having a clear, well understood mission at the district-level, a common vision for improvement, teacher empowerment in terms of input on school-level decisions, a collaborative collegial school culture, and strong leadership were the conditions that most contributed to an individual teacher’s and organizational learning in a school (Leithwood et al., 1998; Mark & Louis, 1999). Finnegan and Daly (2012) argue that high-performing schools provide forums for exchanging knowledge and opportunities to discuss assumptions about teaching and learning. Consequently, in order to improve, low-performing schools must also become learning organizations that provide forums for teachers to exchange their knowledge (Finnegan & Daly, 2012; Higgins et al., 2012). Theory suggests that incorporating collaborative organizational learning would allow teachers to work smarter not harder (Hargreaves, 2001). 21 Social Capital Related to organizational learning, but focusing more on individuals and relationships, social capital theory argues that while people are rational in their choices and actions, they also are guided by their socialization and influenced by social norms, rules, and obligations (Coleman, 2000). Social capital can be defined as the resources available to actors through their relationships and their location in the structure of their relations. These resources can be used to accomplish actors’ interests (Adler & Kwon, 2002; Coleman, 2000). Social capital encompasses both the network of relations and the assets that can be mobilized through the network (Adler & Kwon, 2002; Nahapiet & Ghosal, 1998). Social capital is jointly owned by both individuals in the relationship and is affected by the evolution of social relationships (Nahapiet & Ghosal, 1998). In schools, Baker-Doyle and Yoon (2010) distinguish the social capital within a teacher’s social network as “practitioner-based social capital.” Practitioner-based social capital is the resources, information, and support for effective teaching available in a teacher’s social relationships (Baker-Doyle & Yoon, 2010). Social capital can enhance individual outcomes by increasing access to information through one’s relations, exerting influence over others in the network, improving social credentials through ties with influential people, and reinforcing a sense of identity with a group (Adler & Kwon, 2002; Lin, 2002). On the other hand, there are potential risks in social capital because building and maintaining the relationships necessary for social capital can be costly, strong group solidarity can impede the introduction of new information, and strong identification with a subgroup may compromise identification with the whole and contribute to fragmentation (Adler & Kwon, 2002). 22 Nahapiet and Ghosal (1998) define three dimensions of organizational social capital: structural, cognitive, and relational. The structural dimension of social capital describes the properties of the network of relationships or ties between people or units (Nahapiet & Ghosal, 1998). The structural dimension of social capital describes the ties between actors in a network as strong or weak. Strong ties between individuals facilitate deep and rich communication, while weak ties enable the exchange of information between a more diverse set of actors or between different subgroups in a network (Mohrman, Tenaski, & Mohrman, 2003). The structural dimension also describes the structure of the social network (Nahapiet & Ghosal, 1998), which will be explored in more detail in the next section. The cognitive dimension of social capital refers to the resources that can be shared between people and the shared representations, interpretations, and meanings among people in the same social system (Nahapiet & Ghosal, 1998). The relational dimension of social capital describes the kinds of personal relationships that people have with each other which can be described through shared norms, identities, and obligations as well as feelings of trust (Nahapiet & Ghosal, 1998). Social Networks In order to study the effects of social capital and organizational learning on organizational outcomes, researchers tend to analyze the structure of social networks within an organization. Social network analysis studies both the organization’s structural social capital and an individual’s location within that structure (Carolan, 2013). Social network theory suggests that the structure of social relationships within an organization influences the direction, speed, and depth of organizational change and make transparent the social forces in the organization may support or constrain change efforts (Daly, Moolenaar, Bolivar, & Burke, 2009). 23 According to social capital theory, people rely on their networks of relationships to access information. The social capital available to each member of an organization depends upon the organizational structure of the social network and that member’s location within the social structure (Coburn & Russell, 2008). Organizational members with frequent social interactions to other members with similar beliefs are more willing to share their knowledge and feel more social pressure to do so (Chow & Chan, 2008). The problem is that it can be difficult to get people with different expertise, backgrounds, and mannerisms to effectively integrate their perspectives (Cross et al., 2012). In a fragmented network, characterized by dense subgroups with little interactions across groups, the exchange of knowledge can by slowed or confined to individual subgroups (Atteberry & Bryk, 2010). Researchers distinguish between different types of social networks. The most commonly researched are the instrumental and expressive networks in an organization (Moolenaar et al., 2012). The instrumental network refers to the network of people that individuals go to for advice to achieve organizational goals and work information. The expressive network represents the structure of ties of people that members go to for emotional support, such as friendship, social support and personal advice (Moolenaar et al., 2012). Importantly, researchers have found that a teacher’s expressive network tends to exert a strong influence on her attitude toward new school reform (Cole & Weinbaum, 2010). Studying faculty networks in schools can help produce a better understanding of the internal structure of school community and be used to analyze the flow of resources and expertise through ties in the school (Penuel, Riel, Krause, & Frank, 2009). When designing organizational interventions, it can help to identify the influential people in an organization who may control the flow of information (those central or that fill structural holes) as well as help to 24 recognize what people or departments are peripheral in the network, so that interventions can be made to better connect them with the collective (Cross, Brogatti, & Parker, 2012). In schools, grade levels and department tend to be the main organizational unit in social networks. In studying three schools in San Diego, Daly and colleagues (2010) found that teachers indicated significantly more relational ties with their colleagues in the same grade-level subgroups than with teachers of other grades and support staff. Additionally, the schools’ social networks tended to have high centralization scores as teachers predominately sought out principals for information over other school staff. Social network analysis uncovers the extent to which teachers have access to the resources and knowledge necessary to overcome problems in their daily practice (Moolenaar, 2012). The quality and configuration of ties between teachers, and their access to expertise through their social ties allow for the search and use of knowledge (Coburn & Russell, 2008). Teachers who hold a central position in the school’s network have increased access to the resources and knowledge from a larger range of subgroups (Mohrman et al., 2003). In contrast, isolated members of the school need to be incorporated into the school’s social activity in order to gain exposure to school’s knowledge, resources and practices. Isolated actors also lack the opportunity to share their knowledge with the organization (Daly et al., 2010). Social Networks and Organizational Learning Certain social network characteristics may better support the processes of organizational learning than others (Mohrman et al., 2003; Tensaki & Chesmore, 2003). Dense units, characterized by frequent and common exchange of work-related information, have the ability to better exchange and develop the new knowledge necessary for organizational learning (Moolenaar & Sleegers, 2010). In one study, Mohrman and colleagues (2003) analysis of eight 25 organizations find that social networks make a substantial difference in the capacity of an organization to implement fundamental change throughout the organization, since as they argue, change is a collective process embedded in social communities. They find that hierarchical control does not contribute to the spread of new practices, because these types of relationships limit the two-way knowledge sharing needed to combine and create common knowledge. For organizational change to occur lateral ties are crucial. Lateral ties within departments allow for learning and implementation, while lateral ties between departments allow organizational members to understand the bigger picture of the organizational change (Mohrman, Tensaki, & Mohrman, 2009). Social Networks and School Reform The success of school improvement initiatives is not solely dependent on the knowledge of individual teachers, but also on the extent to which each teachers’ social network allows for the exchange, discussion, and modification of information (Moolenaar, 2012). Social network analysis shows that any reform introduced into schools goes through several layers of modification prior to changing behavior or practice at the classroom level. Scholars describe this process as “the reform is first interpreted by the principal, modified at the grade level, and then finally delivered at the classroom” (Daly et al., 2010, p. 375). Some network structures better facilitate the distribution of information throughout the organization than others (McGrath & Krackhardt, 2003; Frank, Penuel, & Krause, 2015). McGrath and Krackhardt (2003) argue that depending on the characteristics of the proposed change different network structures may better support its organization-wide implementation. The researchers find that in schools where change requires cooperation and trust, more ties across subunits facilitate the conditions necessary for the implementation of the innovation 26 across the school (McGrath & Krackhardt, 2003). In a divided social network, with little interaction among teachers, diffusion of practices and innovations may be slowed or confined to subgroups (Atteberry & Bryk, 2010). On the other hand, in a simulation test, researchers find that when the innovation is controversial, strong interconnecting ties tend to advantage the status quo. When the change initiative is more controversial, it is more successfully implemented across the organization when interaction between subunits is minimal and the change is allowed to develop in a peripheral subgroup, and then slowly taken up by the rest of the organization. Lastly, when an innovation is clearly superior, change is likely to result from the diffusion of information about the change to the friends of randomly selected people, since an individual’s friends are more likely to have more friends than the original individual (McGrath & Krackhardt, 2003). More specifically in school settings, in an analysis of teacher surveys from schools engaged in a reform initiative, Frank et al. (2015) found that schools are better able to implement change when only a few subgroups are responsible for providing the know-how to the rest of the school. This is because a high-implementing subgroup can work together to make the know-how more explicit and consistent before they engage others in the school. This allows for better coordination of the change as it expands across the school. Like any other school policy, school turnaround reforms would be affected by as well as influence the social networks within schools. Low-performing schools tend to have more fragmented networks than higher-performing schools (Finnegan & Daly, 2012). This may make it harder for any school-wide interventions associated with the turnaround to diffuse throughout the organization. Additionally, the replacement of a large proportion of the staff in the turnaround efforts recently mandated by the federal government may result in an even more 27 fragmented social network as teachers work to reestablish their lost professional relationships (Rice & Malen, 2003). Conceptual Framework I draw on improvement science as the conceptual framework for this study. Below, I explore how improvement science differs from how traditional education research, also known as experimental science, conceptualizes school improvement. Additionally, I describe how improvement science incorporates the tenets of organizational learning, social capital, and social networks into its assumptions of change. The purpose of the Blueprint for Systemic Reconfiguration is to re-structure districts into learning organizations that focus on high-quality instruction and high levels of student learning (Frank & Chandler, 2016). Improvement science methods provide a framework for understanding the theories of action and assumptions of change embedded within the Blueprint for Systemic Reconfiguration system as well as in the beliefs and actions of school leaders. Bryk and colleagues (2015) define improvement science as “a methodology that disciplines inquiries to improve practice” (p. 10). Improvement science theorizes that there are two types of knowledge necessary to affect change: basic knowledge and a system of profound knowledge. Basic knowledge refers to knowledge about the topic one is hoping to change. The system of profound knowledge addresses the knowledge necessary to enact the basic knowledge within an organization. This includes a knowledge of systems, knowledge of variation, knowledge of psychology, and knowledge of how learning happens in organization (Lewis, 2015). Improvement science provides tools and processes for combining these two streams of knowledge in order to affect change in an organization. Overall the main theme of improvement science research is “learning through doing (Bryk, Gomez, & Grunow, 2011). Improvement 28 science aims to answer the question “what works, for whom, and under what set of conditions?” (p. 14). It is based on researchers and participants jointly engaging in multiple cycles of improvement design, testing, and implementation (Bryk et al., 2011). In education, improvement science studies how to improve the capacity of school organizations to increase valued outcomes for all students regardless of their teacher or school context. It rests on the assumption that the key to improving educational results is not “getting teachers to work harder; it’s about getting them to work smarter” (p. 27). To get teachers to work smarter, Bryk and colleagues (2015) outline processes that ask schools to “learn fast to implement well.” These processes form the foundation of improvement science. Traditional Education Research Science – Experimental Science Prior to the introduction of improvement science methods into education, the passage of No Child Left Behind (NCLB) gave preference to “scientifically based research.” In fact, the law used the phrase 110 times (Slavin, 2002). NCLB defines scientifically based as “rigorous, systematic, and objective procedures to obtain valid knowledge.” The phrase “scientifically- based” research is understood as research that uses experimental or quasi-experimental designs with the hallmark method of randomized control trials (RCTs) (Lewis, 2015; Slavin, 2002). Simply put this research approach focuses primarily on questions of “what works” (Peterson, 2016). Experimental science methods prioritize causal attribution. This priority requires researchers to put constraints into their research – such as a singular treatment, and controls, such as eliminating exogenous factors – in order to make these inferences (LeMathieu, Edwards, & Gomez, 2015). Slavin (2002) argues that these “rigorous research” standards have the potential to dramatically shift the use and prestige of educational research. Research completed to these 29 standards can prove positive effects of programs, which can then be replicated through replication paradigms that define development, rigorous evaluation, replication, and dissemination. Slavin emphasizes that this type of research allows for findings and causal conclusions that will allow educators and policymakers to ask the question, “if we implement program X instead of program Y, or instead of our current program, what will be the likely outcomes for children?” (p. 18). Overall, Slavin asserts that this type of research will allow for research results that are useful to educators and policymakers, which allows for the potential for evidence-based policies to be implemented which could transform educational practices (Slavin, 2002). This tradition of research assumes the tested program’s positive effects occur through the scaling up of faithful implementation of the proven program in new settings, regardless of local context (Bryk, Gomez, & Grunow, 2011; Lewis, 2015) Problems with Traditional Research Science. Simply, context matters especially when attempting to implement complex ideas into practice, such as high-quality instructional practices. As summarized by LeMathieu and colleagues (2015), experimental science “helps us to know that some practice can produce some effect, it does little to show us how to produce those effects over and over and across people and places.” (p. 446). RCTs produce evidence relevant only to single either-or decisions, such as which program to implement (Peterson, 2016). LeMathieu and colleagues (2015) argue that these methodological requirements limit the ability of the research’s ability to produce research that can inform practice in real-world settings. Put in other words, RCTs and their research techniques ignore the interdependence of factors that may affect the efficacy of programs in differing contexts (Peterson, 2016). Experimental science and its RCTs may be an overly simplistic research design when variation in setting is likely to affect 30 implementation or when will and capacity need to be built, such as in educational settings (Lewis, 2015). The priority of ‘what works’ studies is to provide an estimate of the likely impact of an intervention when implemented at scale (Peterson, 2016). In other words, the experimental research paradigm places its focus on determining “cause” without considering the applicability of the findings to different people in different places and under different circumstances (Bryk, Gomez, & Grunow, 2011) These ‘what works’ studies fail to address how, why, or under what conditions the program works. The result is very little information that can be used by practitioners to inform their day-to-day work, as they focus on practices instead of programs (Cannata, Cohen-Vogel, & Sorum, 2017; Peterson, 2016). Components of Improvement Science Improvement science brings a scientific lens to the context of human experience in the real world (Perla, Provost, & Perry, 2013). Improvement science “reminds us that there is usually no single magic bullet solution to a problem and that the cycle of discovery and justification is best viewed as an iterative (and continuous) process” (Perla et al., 2013, p. 179). Instead, improvement science acknowledges that improving at scale requires productively coping with local diversity in people, places, and circumstances (Bryk et al., 2011). At the foundation of improvement science is the idea that improvement comes from the process of developing, testing, and implementing changes (Lewis, 2015). Improvement science shifts the notion of scaling up from faithful implementation to adaptive integration. Adaptive integration involves using improvement science processes to learn what it takes to make an innovation or change that worked in one context work in another situation (Bryk et al., 2015). 31 This process of improvement science assumes that scale-up of changes occurs through the integration of basic knowledge, such as content knowledge, with profound knowledge. Profound knowledge describes the knowledge about how to build shared ownership of improvement across the organization, to detect and learn from variations in practice, and to share knowledge among practitioners (Lewis, 2015). Profound knowledge also describes the operational definitions common to those in the organization. Operational definitions, similar to the concept of informal distribution from organizational learning, provide a shared meaning and understanding of concepts, ideas, goals, and measures. “Without operational definitions, the meaning and intent of actions and words are known only by individuals who use them” (Perla et al., 2013, p. 179). This focus on profound knowledge and operational definitions highlight the importance of effective communication and shared understanding for collective action throughout the organization around a change effort (Lewis, 2015; Perla et al., 2013). Overall, improvement science processes address a challenge with current reform efforts and what Bryk, Gomez, and Grunow (2011) refer to as the “natural tendency to grasp for promising solutions or best practices without fully understanding how such solutions must be integrated with other solutions and preexisting organizational conditions” through a focus on systems-thinking (p. 139). Systems are defined as a network of interdependent components that work together to accomplish the aim of the system. People are one part of systems along with policies, processes, organizational structures, operating rules and culture. This means that systems-thinking must focus on how these interdependent components relate to one another and work together. The design of the components of the systems determine the results that the system achieves (Perla et al., 2013). Further, Perla and colleagues (2013) describe that “systems thinking means viewing an organization as dynamic, adaptive to the needs of the customer, and composed 32 of interdependent people, departments, equipment, facilities, processes, and products, all working toward a common purpose” (p. 182). Importantly, this quote stresses that systems are dynamic and complex making it difficult to uncover any delayed effects of change interventions and any unintended consequences (Perla et al., 2013). Langley and colleagues (2009) outline five central principles of improvement science: (1) knowing why you need to improve, (2) having a way to get feedback to let you know if improvement is happening, (3) developing a change that you think will result in improvement, (4) testing a change before any attempts to implement at scale, and (5) implementing a change (p. 17-18). Knowing why you need to improve. This principle refers to defining the aim or purpose of the improvement effort (Langley et al., 2009). It stresses that the targeted work in improvement science must be problem-specific and user-centered. In order to achieve this focus leaders employing improvement science methods to scale up changes must consider multiple perspectives and most importantly subject matter experts, who have experience at the implementation level (LeMathieu et al., 2015; Lewis, 2015). Having a way to get feedback to let you know if improvement is happening. This principle addresses the need to have a feedback loop in place, so that leaders will know if the improvement is working (Langley et al., 2009). In many cases defining the feedback loop involves putting in place some sort of data collection system. Data collection processes must be embedded into daily work, rather than added on top of the day-to-day work of program participants (Bryk et al., 2011). This principle highlights that system dynamics is the consequence of feedback mechanisms, so it is essential to define adequate systems to inform decisions and future work (Perla et al., 2013). 33 Developing a change that you think will result in improvement. Developing a change plan requires input from a variety of stakeholders (Langley et al., 2009). Effective changes must be informed by the experience, knowledge, and intuitions of subject matter experts (Lewis, 2015, Perla et al., 2013). The prior experiences of subject matter experts form the mental models on which the predictions of any change effort are formed (Perla et al., 2013). Effective changes are those that are customized to the local setting instead of a list of tasks that are assumed to be generalizable to any setting (Perla et al., 2013). Any change idea must be specific enough to be able to be made and sustained (Langley et al, 2009). Testing a change before any attempts to implement at scale. Improvement science is grounded in testing and learning cycles (Perla et al., 2013). Langley and colleagues (2009) describe a process known as plan-do-study-act cycles (PDSA cycles) in order to test changes. This process has four steps. First, change agents must plan the test by defining who, where, when, what, and how the test will occur including what information or data will be collected. Second, they must put the test into action and run the test. Change agents should be open to learn about both the planned and unexpected results of their change. The third step is to summarize what knowledge was gained from the test, so that in the fourth step whichever action is decided to be taken is based on information gained from the test. The action decision may include implementing the change, abandoning the change, or refining the change and testing again (Langley et al., 2009). PDSA cycles justify whether the change worked using some sort of data. These cycles answer both “what do we know” and “how do we know” (Perla et al., 2013). In improvement science, a negative test result is not to be interpreted as a failure but instead as more information for the following test (Perla et al., 2013). 34 A major principle of improvement science is that variation is helpful. Bryk and colleagues (2015) argue that “understanding the sources of variations in outcomes, and responding effectively to them, lies at the heart of quality improvement” (p. 35). Improvement science considers the middle ground between standardization and pure flexibility in terms of professional work. “It acknowledges that variability in performance is a major problem to solve, while also affirming that individual professional judgement will continue to play a central role in practice” (Bryk et al., 2015, p. 50). Moreover, this variation caused from professional judgement that resulted in especially positive results represents opportunities to learn about what actions might lead to more improvement in some contexts than others, which offers insights into the sources of improvement (Peterson, 2016). This variation can provide valuable information during PDSA cycles for researchers and practitioners to use to refine their work processes (Bryk et al. 2015; Langley et al., 2009). Implementing a change. This component refers not to the questions from experimental science around the quality of the change, but instead refers to the process of making the change permanent. This is not a mechanistic process, because people are involved. People have their own motivations, beliefs, and capacity for implementing a change. Consequently, when implementing a change, this component stresses the importance of making the rationale for the improvement effort “clear to everyone involved.” (Langley et al., 2009, p. 21). This process of making clear the rationale can be achieved by inviting people early in the process to participate in the development and testing of the considered change. (Langley et al., 2009). 35 Summary Figure 2.1 from Lewis (2015, p. 56) provides a summarized comparison of the difference between the experimental science paradigm and the improvement science paradigm in terms of scaling up instructional strategies to build students’ mathematical strategies. Figure 2.1: Scale-up of Knowledge: Contrast of Paradigms (from Lewis, 2015, p. 56) 36 Improvement Science in Education Schools and school districts offer a promising system for improvement science methods. As described above in terms of social networks, in education, knowledge is often localized and consequently difficult to transfer. This means that improving at scale will require responding to local diversity (Lewis, 2015). Improvement science methods align with the daily context of teaching and learning (Peterson, 2016). As described above, improvement science combines two different streams of knowledge: basic knowledge and system of profound knowledge (Lewis, 2015). In educational terms, basic knowledge is from the discipline of education, for example knowledge about mathematical instructional strategies. While the system of profound knowledge is defined as the knowledge of systems, variation, and knowledge of how knowledge grows. In other words, it describes knowledge of improvement science principles (Lewis, 2015). There is very little educational research in the improvement science tradition because it is relatively new (Lewis, 2015). However, currently there are two developing streams of improvement science research in education settings: networked improvement communities and lesson study. Networked Improvement Communities. Drawing on the social capital and organizational learning theories described above, in the book Learning to Improve, Bryk and colleagues (2015) propose an improvement science model of organizational change. The authors describe the creation of networked improvement communities (NICs) to collectively develop and implement organizational change such as innovations in instructional practice. NICs are intentionally designed social organizations organized around a deep understanding of a problem and a shared working theory to improve it. These communities engage diverse perspectives from members of many organizations organized around a coordinating hub. The work of a NIC 37 includes defining the causes of their common problem, developing possible interventions that address their theories of change, and attempting these interventions in the represented organizations. These communities work to break down the silos between individuals and between organizations that limit the sharing of practical knowledge and positive improvements. The intention for NICs is not to design an intervention that works universally, but instead to determine “what works, when, for whom, and under what sets of circumstances” (Bryk et al., 2011, p. 23). The authors propose a process that develops learning in the network and transfers this knowledge from the individual to the organization to the community of organizations. This learning process starts with the definition of a focused, common aim. Community members then enact the plan-do-study-act (PDSA) cycle in order to guide their learning in their improvement efforts, as described above as a component of improvement science research (Bryk et al., 2015, p. 122). Recent research on NICs has shown that nearly all participants working to improve student outcomes through PDSA cycles expressed a belief in the value in the cycles; however, researchers found mixed levels of enthusiasm for actually conducting PDSA cycles as school officials experienced time and data collection constraints (Tichnor-Wagner, Wachen, Cannata, & Cohen-Vogel, 2017). These PDSA cycles in NICs bring together the perspectives of a diverse group of organizational and network community members in deep conversations that allow for the construction of common definitions of the problems and solutions, like the information interpretation process in organizational learning (Schechter & Qadach, 2012). These networked communities engaging in PDSA cycles allow for the combination of individual experiences and 38 knowledge about the organization and the change in order to create a more complete common knowledge about improving the organization. BTEN network. An example of a networked improvement community working in schools is the Building a Teaching Effectiveness Network (BTEN) (Hannan, Russell, Takahashi, & Park, 2015). The BTEN worked for three years to improve systems of support for early career teachers. In the first year, researchers worked with network members from one elementary school to develop an observation feedback process to help principals provide meaningful and timely feedback to novice teachers. BTEN began by defining elements of new teacher support: the development of trusting relationships with administrators and the reliable delivery of context- specific, high-quality, and actionable feedback. These definitions resulted in work to improve the feedback process for new teachers with the hope of ameliorating the problem of new teacher attrition in partnering schools (Hannan et al., 2015). In years 2 and 3, the BTEN added 5 schools and 17 schools in each respective year that they charge with integrating the feedback process into their school organizations. This number of schools allowed researchers and BTEN members to analyze the variation in implementation of the feedback process to learn about how context, capacity, and systems influence the efficacy of the feedback process. Researchers working with the BTENs categorized their partnering schools by the extent to which they used improvement science methods to learn about supporting beginning teachers and used this learning to modify and implement a new feedback process into their school. Of the six focal schools, two schools fell into each category: incomplete enactment schools, learning en route to action schools, and the systemic knowledge into practice schools. The two schools with the highest improvement science practices, in the systemic knowledge into practice category, both adopted the feedback process with integrity by using improvement science principles to 39 adapt the standard feedback process to match each school’s context. For example, at Redwood, they added an intermediate goal check-in step between the initial principal meeting and the teacher’s work (Hannan et al., 2015). Schools at the intermediate level, learning en route to action schools, used improvement science methods but experienced complications when attempting to apply the knowledge they gained through improvement science methods to adaptively integrating the feedback process. The last two schools, incomplete enactment schools, struggled to enact improvement science processes. This may be due to a lack of clearly defined roles, high personnel turnover, and multiple, competing initiatives. Overall, Hannan and colleagues (2015) demonstrate both the promise of improvement science methods used in schools to improve feedback processes for new teachers as well as some difficulties in incorporating improvement science methods into school processes. In the Blueprint, the work of the district network and building networks mirrors the literature on networked improvement communities (Bryk et al., 2015). Like NICs, the Blueprint’s networks engage a diverse group of stakeholders from around the district and school building to exchange their knowledge and perspectives. The networks work to develop new organizational protocols and practices aligned with the pre-established Blueprint system, try them out in the organization, and then come together to adjust them to fit to their organizational needs. Through the use of networks employing practices similar to those in the networked improvement communities literature, districts strive to collaboratively construct a district vision of high-quality instruction, implement it throughout all district classrooms, and respond to obstacles in implementation at both building network and district network meetings (Frank & Chandler 2016). 40 Lesson Study. Another form of improvement science methods in education is lesson study, which began in Japan and was brought to the United States by Stigler and Hiebert (1999) in their study of the practices of countries in the Third International Mathematics and Science Study (Lewis, Perry, Friedkin & Roth, 2012). Fernandez (2002) notes “Lesson study is more than the study of lessons; it is rather a systematic inquiry into teaching practice much more broadly defined, which happens to be carried out by examining lessons” (p. 394). Lesson study begins when a team of 3-4 teachers set a goal for their students that they want to address in their upcoming lessons. The team then comes together to meticulously plan the lesson collaboratively creating a written lesson plan. Then one of the team’s teachers teaches the group-designed lesson to his/her students while the other team members observe the lesson and take notes (Fernandez, 2002). After the observation, the team reconvenes for the teachers to share their observations of the lesson and work to create a modified version of the lesson plan that addresses any changes that the team deems important based on the classroom observation. With the revised lesson plan, a second teacher from the group teaches the new version to his/her students while the other teachers again observe. Afterwards, the teachers meet again to share their comments and suggestions from the observation (Fernandez, 2002). These observed lessons also known as “research lessons,” act as a shared vision of education common to the group (Lewis, Perry, & Murata, 2006). Finally, lesson study teams periodically write summary reports to provide a record of the insights that the group has made to be shared with the larger profession. Typically, a lesson study cycle takes 10-15 hours of group meetings over 3-4 weeks (Fernandez, 2002). 41 Figure 2.2 shows the lesson study process (Lewis et al., 2006, p. 4). It should be noted that the steps are similar to those of a PDSA cycle completed by NICs. Figure 2.2: Lesson Study Process (from Lewis et al., 2006, p. 4) Lesson study also is based mainly on the assumptions of the situated learning perspective, which like social capital and organizational learning theories, understands that learning is situation-based and collaborative. Teachers work in lesson study teams creating a professional community of inquiry with their colleagues (Perry & Lewis, 2009). The inquiry cycle of lesson study begins as team members collectively establish a goal and collaboratively design a lesson plan (Fernandez, 2002). The observations of other teachers allow team members to see each other’s practice and develop shared meaning of high-leverage practices, such as “scaffolding” (Lewis et. al., 2012). Lewis and colleagues (2012) observed, “In a lesson study community, it is 42 not unusual for teachers to take initiative to build, test, refine, and spread instructional improvements trough their collegial networks” (p. 372). Beyond building common knowledge, collaboration achieved in lesson study can also build shared professional norms and motivation (Lewis et. al., 2012). In Japan as lesson study has become commonplace and knowledge disseminates, the knowledge around lessons is developed as they are taught in different local contexts across the country. This allows lessons to be adapted to local structures, studied for effectiveness, and ultimately improved in design and theory (Lewis et al., 2006). In a more specific case, in their research of the use of lesson study for school turnaround, Saito and Sato (2012) found that by using the structure of lesson study to focus on the students, teachers learned more information about the realties of children and began to notice the verbal and non-verbal signals of their students concerning their emotions, level of understanding, and their relationships. Teachers at the turnaround school used lesson study to design lessons that allowed students to work in groups, share their opinions, and feel a sense of belonging in the class. As a result, teachers better met the needs of their students (Saito & Sato, 2012). The Usefulness of Improvement Science for Analyzing School Improvement Efforts Previous research raises questions about the ability of policy to influence instructional practices at scale. School turnaround policies over the past 20 years have overlooked the potential of districts to support school improvement efforts as well as the importance of teacher social capital and social networks in supporting teachers’ interpretation of new policy. Improvement science brings together these concepts into a framework for scaling up that is new to the education sector. Put differently, Bryk and colleagues (2011) note that improvement science research: 43 focuses our attention on the information necessary to make interventions work reliably at scale. Rather than thinking about a tool, routine, or some other instructional resource as having proven effectiveness, improvement research directs efforts to understanding how such artifacts can be adaptively integrated with efficacy into varied contexts, for different kinds of students, and for use by diverse faculty (p. 149) Improvement science and networked improvement communities could provide a social learning alternative to the current turnaround policies. This alternative capitalizes on the local knowledge already existing in the organization and provides the necessary structures and routines to promote organizational learning (Schechter & Qadach, 2012), as well as the creation of social ties and social capital that will allow for the access of knowledge in similar organizations attempting the same changes (Adler & Kwon, 2002). This dissertation studies the Blueprint for Systemic Reconfiguration, which intentionally designs systems at the district- and school-levels that target the improvement of teacher instructional practice and student learning at scale across the organization. As described above, the Blueprint uses nested NIC-like networks that design and test change ideas in district and school settings. In this dissertation, I study how district and school leaders use and do not use the assumptions of change embedded within the improvement science paradigm to work to turnaround student performance in reading though influencing teacher practices. More specifically, I study how the operational definition of high-quality teaching, as represented by the reading subject-specific vision for high-quality teaching, moves through district systems and ultimately influences instructional practices. 44 CHAPTER 3: RESEARCH DESIGN Research Design and Research Questions This dissertation strives to understand how stakeholders’ assumptions of change can affect change in the instructional core of schooling at scale throughout the district through the implementation of the Blueprint for Systemic Reconfiguration. More specifically, I examine: each district’s theory of change, how that theory is implemented in 1. What is the intended theory of change of the Blueprint for Systemic Reconfiguration for influencing instruction at scale? 2. How does each administrator’s orientation toward change influence the implementation of Blueprint for Systemic Reconfiguration in their district or school? 3. How is the theory of change enacted in each district? How does this enactment reflect administrators’ orientations toward change? 4. To what degree and in what ways does the enactment of the theory of change influence teacher instructional practices and planning in each school? I used an embedded multiple case study methodology to examine these issues (Yin, 2014). This dissertation examined two districts in their first full school year of implementing the “Blueprint for Strategic Reconfiguration.” Within each district, I focused on two elementary schools and three teachers within each school, for a total of 12 teachers across the two districts. The teachers were chosen to maximize variation in teacher involvement in leadership activities (Patton, 2002). At each school, at least one was a member of the building network and at least one teacher was not a member of the building network and did not have any other leadership roles. This design of including two districts, four schools, and 12 classrooms allowed me to study 45 how the Blueprint for Strategic Reconfiguration worked at each level of the organization to influence the instructional core. Case study research incorporates many variables of interest, relies on multiple sources of evidence, and incorporates prior theoretical propositions to guide data collection and analysis (Yin, 2014). Case study research allows researchers to place themselves within the study context and understand the viewpoints and behaviors of social actors (Flyvbjerg, 2006). Yin (2014) asserts that case study is an appropriate method when the following three conditions are met: “(a) the type of research question posed is a how or why question; (b) the researcher cannot manipulate the behavior of study participants; and (c) the study focuses on contemporary events” (p. 9). As the proposed study meets all of these conditions, case study methodology is an appropriate design choice. Case study research can take three different forms: exploratory, descriptive, and explanatory. First, exploratory case studies strive to develop research questions, hypotheses, or procedures that can be used in future research. In contrast, a descriptive case study strives to describe a phenomenon in a real-world context. Lastly, an explanatory case study seeks to explain how or why some condition came to be (Yin, 2014). This dissertation research was an explanatory case study since it strived to explain how district and school leaders’ theories around change working in an improvement science framework could change instructional practice and why this change influenced instruction in some classrooms more than others. These explanations have the potential to inform hypotheses on how local policies and procedures do or do not have the potential to influence instructional practices at scale throughout school districts. To be able to compare theories of action and changes in the instructional core across different schools, I used an embedded multiple-case design (see Figure 3.1). Yin (2014) 46 recommends a multiple case study design over a single case because multiple case studies allow for analytic conclusions to originate from two contexts rather than a singular context. When case study researchers complete single-case studies there can be criticisms and fears that any findings are unique to the specific context of that case. Having two or more cases can work to dull these fears (Yin, 2014). CONTEXT CONTEXT Case Hillside School District Embedded Unit of Analysis Hickory Elementary School Teacher Teacher Teacher A B C Case Lakeside School District Embedded Unit of Analysis Cottonwood Elementary School Teacher Teacher Teacher A B C Embedded Unit of Analysis Willow Elementary School Teacher Teacher Teacher A B C Embedded Unit of Analysis Rose Elementary School Teacher Teacher Teacher A B C Figure 3.1: Design of Case Study (adapted from Yin, 2014, p. 50) Participants and Sampling Strategy Yin (2014) recommends that multiple case study designs use replication logic when selecting possible cases. Replication sampling logic calls on case study researchers to identify cases that either predict similar results (literal replication) or predict contrasting results 47 (theoretical replication). This dissertation design identified two similar districts in which one would predict similar consequences of implementing the Blueprint for Strategic Reconfiguration. I chose two suburban districts near a large metropolitan center in Michigan. These districts were recommended by the director of the Blueprint for Strategic Reconfiguration, Dr. Chandler, as districts that were progressing well in the planning period and preparing to launch the systems district-wide. Both districts were in the same geographical area. The first district, Hillside Community Schools1, served just over 3,000 students. A racially diverse student body attended Hillside, where approximately 60% of students were white, 3% were Hispanic/Latino, 35% were African American, 2% were two or more races, and less than 1% of students were Asian or Native American. Additionally, 70% of the district’s students were economically disadvantaged. Hillside School District employed approximately 220 teachers. In the study year, 25% of the teachers were in their first five years of teaching. The second district, Lakeside Community Schools, served just under 5,000 students. In Lakeside, approximately 45% of the students were white, 40% were Hispanic/ Latino, 10% were African American, 5% were of two or more races, and less than 1% of students were Asian or Native American. Approximately 80% of students were economically disadvantaged. Lakeside School District employed approximately 310 teachers. During the study year, 36% of them were in their first five years of teaching. In terms of the Blueprint, both districts were in their second year of working with the Blueprint. The second year involved implementing most Blueprint systems (see Appendix A for the Blueprint installation timeline). 1 The names of the districts, schools, and all participating individuals are pseudonyms, with the exception of Dr. Grant Chandler, the director of Blueprint for Strategic Reconfiguration. Identifying information about the districts and schools have been altered slightly to protect their anonymity. 48 Within the two districts, I decided to focus on the elementary schools. Focusing only on elementary schools allowed me to further the replication logic by comparing findings in one elementary school to another elementary school in which teachers would be teaching similarly aged children and similar curricula. At Hillside Community Schools, I focused on Hickory and Willow Elementary Schools. Hickory and Willow Elementary Schools each served approximately 300 students from kindergarten through fifth grade. At Lakeside Community Schools, I focused on Cottonwood and Rose Elementary School. Cottonwood and Rose Elementary Schools each served approximately 300 students in Pre-K through fifth grade. Additionally, Rose Elementary School was identified as a priority school by the State of Michigan on the 2015 list for placing in the bottom 5% of schools during the 2014-2015 school year. During the study year, 2016-2017, Cottonwood Elementary was recognized as a Rewards School for the 2015-2016 school year due to student growth and Rose Elementary was released from priority status. Within each elementary school, I recruited three teachers to maximize variation in membership on leadership teams (Patton, 2002). At least one was a member of the Building Network (BN), and at least one was not a member of the BN and did not have any other leadership roles under the Blueprint. Additionally, at Hillside School District, at least one teacher participated in the curriculum writing process at the district level. To recruit these teachers, I asked school and district leaders to identify teachers who met these categories and would be open to an outside observer. School leaders identified these teachers and provided contact information. Only one teacher asked not to participate before the study began and in that case the school principal recruited another teacher to participate. Table 3.1 shows background information on the research participants. 49 School Name Mr. Dixon Ms. Wilson Ms. Parrish Ms. Edmund Ms. Johnson Ms. Reed Ms. Stewart Ms. Clark Ms. Olin Ms. King Ms. Davis Ms. Morey Ms. O’ Neill Ms. Miller Ms. Nelson Ms. Trainor Ms. DeSutter Ms. Lavin Table 3.1: Participant Information District Lakeside Lakeside Cottonwood Lakeside Cottonwood Lakeside Cottonwood Lakeside Cottonwood Lakeside Lakeside Lakeside Lakeside Hillside Hillside Hillside Hillside Hillside Hillside Hillside Hillside Hillside Rose Rose Rose Rose Hickory Hickory Hickory Hickory Hickory Willow Willow Willow Position Superintendent Principal Teacher Teacher Teacher Principal Teacher Teacher Teacher Superintendent Principal Teacher Teacher Teacher Principal Teacher Teacher Teacher Grade Years of Experience Member of DN Member of BN Other District Leadership Role 5th 3rd 2nd 3rd 4th 5th K 5th 2nd 2nd 2nd 4th 26 14 30 10 10 10 0 24 0 10 29 13 X X X X X N/A X X X X X X N/A X X X Piloting Reading in Social Studies/ Science No Served as Reading Instructional Coach No No Part of original group at MI Excel Training No ELA Curriculum Writer for 5th Grade No Joined Math Curriculum in March Joined Math Curriculum in March No 50 Data Collection Yin (2014) describes six sources of evidence commonly used in case study research: documentation, archival records, interviews, direct observations, participant-observation, and physical artifacts. Additionally, Yin (2014) also suggests that surveys can be a good complement to case study evidence as a way to gain further indications of the prevalence of a phenomenon. Multiple sources of evidence are crucial in case study research because any findings must be confirmed through triangulation (Yin. 2014). These multiple forms of evidence ensure that the phenomenon studied is explored through a variety of lenses that provide differing insights into multiple facets of the phenomenon. The convergence of different types of data sources strengthens case study findings and understanding of the phenomenon (Baxter & Jack, 2008). For this study, I will rely on four types of evidence: (1) direct observations, (2) documentation, (3) interviews, and (4) surveys. Direct Observations MI Excel Training. Over the course of the 2016-2017 school year, I attended Blueprint trainings offered by MI Excel leaders. On September 28-30, 2016, I attended the “Diving Deeper into the Blueprint-Level I” training. The MI Excel website describes this training as an opportunity for participants to “engage in an overview of how to use their knowledge of the Blueprint to support districts in building and installing systems that are sustainable and will have the greatest impact on teacher and student performance.” More specifically this training focused on the District Network, Problem-Solving Driver System, Communications Driver System, and Performance Management Driver System components of the Blueprint. On February 6-8, 2017, I attended the “Diving Deeper into the Blueprint-Level III” training. This level III training focused on the Principals Leverage District Systems, Instructional Leadership 51 Routines, and Teacher Collaborative Routines components of the Blueprint and how to build and install these routines in districts. Prior to attending the level III training, leaders at MI Excel emailed me the PowerPoint slides for the level II training that I was not able to attend. These PowerPoints addressed the Talent Management, Leadership Network, Instructional Infrastructure, and Intense Student Support Network components of the Blueprint. In addition to these training opportunities, I also attended both the 2016 and 2017 Blueprint Institute hosted by MI Excel. The Blueprint Institute was a two-day conference targeted towards Blueprint installing districts that provides an opportunity for districts to share best practices with one another and develop a deeper understanding of the Blueprint framework. At the 2016 institute, I attended sessions on an Introduction to the Blueprint, I attended sessions entitled: (1) Introduction to the Blueprint; (2) Our Story: Traverse City Public Schools; (3) Superintendents’ Panel Discussion; (4) Installing Teacher Collaborative Routines; (5) The Blueprint’s Foundational Elements; and (6) Our Story: Hillside Schools. At the 2017 Blueprint institute, I attended sessions entitled: (1) Our Story: Hillside Schools (2) Our Story: Lakeside Public Schools, (3) Instructional Infrastructure, (4) Superintendents’ Panel Discussion, (5) How do Principals Leverage District Systems, and (6) Installing Instructional Leadership and Teacher Collaborative Routines. At each session, I took field notes and collected documents shared during the presentations. Additionally, as a part of the 2017 Institute all PowerPoint presentations were made available to all participants through a Google Team Drive. Collectively these PowerPoint slides and the notes that I took at the institutes and trainings provided information about the theory of action as conceived by MI Excel leaders for how these systems could work to influence classroom instruction and student achievement. 52 Network Meetings. To address my second and third research questions, I observed district networks and building networks 1-2 times per month over the course of one school year from October 2016 through June 2017. At these meeting, I took running field notes on the discussion relevant to the district’s vision of high-quality instruction and how they planned to influence instructional practices. In my field notes, I took care to note references to beliefs about change processes, what and how information on the implementation process entered the conversation, and how decisions were made. Classroom Instruction. For each focus teacher, I observed four ELA lessons over the school year except for those teachers at Hickory Elementary School. I did not receive contact information from the principal for these teachers until January 2017. Consequently, I only completed 3 observations with these teachers. Additionally, I missed the March observation with Ms. Lavin due to a couple of weeks of school absences in her schedule for a family emergency. During these observations, I looked for evidence of the teachers instructing in ways that correspond with the vision of quality teaching and the instructional framework developed by each district. I recorded my observations through detailed field notes on the actions of the teacher and students during the ELA lesson (see the observation protocol in Appendix B). These classroom observations allowed me to respond to research question four. Documentation District-Made Documents. District- and school-constructed documents regarding changing the instructional practices in the district also informed this study. These documents included information on the district vision of high-quality instruction (e.g. see the MI Excel example vision of high-quality reading instruction in Appendix C), subject specific instructional frameworks, district communications on instructional practices, professional development 53 documents, teacher observation protocols, and other documents relating to instructional practices of teachers in the district. I obtained these documents by periodically checking each district’s website on their Blueprint work, at network meetings, and from focus teachers during their classroom observations. The purpose of collecting these documents was to address the second and third research questions as well as to better understand district plans for influencing the instructional core and the messages that teachers were receiving about how the district expected their instructional practices to change. These documents also allowed me to search for differences in instructional visions and expectations as well as communication patterns between districts and between schools. Interviews In this dissertation, I interviewed the district superintendents, focus school building administrators, and focus teachers multiple times for a total of 56 interviews. These interviews were conducted in one-on-one settings and were audio-recorded. Through the use of semi- structured interviews, I gained an understanding of how the school leaders and teachers were envisioning the changes in instructional practice, their action plan for communicating the expected changes, and whether and how the teachers actually used this information in their planning practices. Interviews can be vital complements to other sources of evidence, such as observations and surveys. Not only can they be used to corroborate quantitative survey findings, but also to extend them with information on participant reasoning about their instructional interactions with colleagues (Spillane, Hopkins, & Sweet, 2015). Superintendent. Twice during the school year, I interviewed each district superintendent. These interviews focused on the district vision of quality instruction, the progress 54 being made on districtwide enactment of that vision, the next steps in promoting the envisioned change, and any obstacles they were facing. The data from the superintendent interviews helped me address the first, second and third research questions. Please see the fall interview protocols in Appendix D. The spring protocol was similar but included new questions based on topics that arose from interacting with teachers, such as the HQI walkthroughs and the use of data in their practice. See Appendix E for the spring superintendent interview protocol. Building Principal. I also completed semi-structured interviews with the school building administrators in the focus schools once in the fall and once in the spring. These interviews addressed the network meetings, how the principal distributed information on instruction to the school’s teachers, and the planning for the upcoming quarter to promote instructional change. Interview data with the building principals helped address research questions 2 and 3. Please see the fall interview protocol in Appendix F. The spring protocol (see Appendix G) was similar but included new questions that arose over the course of data collection, such as around communication strategies, HQI walkthroughs, and their use of data. Focus Teachers. After each classroom observation, I interviewed the observed teacher for a total of four interviews with each teacher, except for with the previously noted teachers at Hickory Elementary and Ms. Lavin. Each interview covered the teacher’s lesson planning practices for the observed lesson, their personal vision of high-quality instruction, the information they had received about the Blueprint and district instructional initiatives, the colleagues that they discussed the Blueprint with, and the reasoning behind the decisions that teachers made. These interviews with the focus teachers helped address research questions 2, 3, and 4. See the interview protocol for the first post-observation in Appendix H. Subsequent interview protocols were similar but questions were added in response to topics that emerged, 55 such as the HQI walkthroughs, data use, and state accountability list information. See appendices I, J, and K for additional interview protocols. Survey In the fall and spring, I also emailed a survey link to the full school staff at each focus school to inquire about their ELA instructional practices. The first administration occurred in October and November 2016 at Lakeside and in January and February 2017 in Hillside. The second administration occurred in both districts in May and June 2017. This survey was shared with all school staff in the focus schools. In the fall, 72.5% of teachers in the four elementary schools completed the survey and in the spring 56.3% of teachers completed the survey. The survey asked about teacher ELA instructional beliefs and practices. Sample questions included “How often do you do the following in your Reading class? – Read aloud to the class” and How often do you have students… compare what they have read with experience they have had?” It also included the questions: “Who do you talk to in your school building about your ELA instruction?” and “Who do you talk to in your district about your ELA instruction?” so that I could explore the instructional advice social network in each school. The survey was piloted in the Fall 2016 course of TE 939D: Qualitative Case Study Methods. This course primarily was attended by doctoral students from the Teacher Education program with approximately 8 students specializing in literacy instruction. These colleagues read the survey for clarity in wording around topics addressing literacy instruction as well as the overall length and flow of the survey. Data from these two administrations of the survey were analyzed to address research question 3. Please see the fall survey in Appendix L. The spring survey addressed similar topics but included some additional items around beliefs of policy coherence (see Appendix M) 56 Quantitative Measures For the quantitative analyses described below, I used data from the previously described surveys to form needed variables and composites. Aligned Teaching. On the survey, teachers responded to at least 29 questions regarding their Reading instruction over the past 5 lessons. Using a subset of these responses, I constructed a measure of aligned teaching based on district documents. For the Lakeside measure, I referred to the Lakeside vision of high-quality reading instruction and the reading implementation guide to identify teaching practices that would align with district expectations. For Hillside, I referred to the Hillside vision of high-quality reading instruction and the reading instructional framework. See Tables 3.2 and 3.3 for the list of teaching practices considered to be aligned to each district’s expectations. 57 During reading instruction, in your past 5 lessons, how often did you...? During reading instruction, in your past 5 lessons, how often did students...? Read aloud to the class Model strategies for students to learn new vocabulary from Ask students to read silently on their own Teach students strategies for decoding sounds and words Ask students questions about texts Form and work with small groups around leveled texts Form and work with small groups around reading skills Read individually Read in small groups Listen to the teacher read Talk with other students about what they have read Read self-selected texts at their independent level Turn and talk Reread a passage in a text Collect evidence from the text using tools such as sticky notes or highlighters Work on individual reading goals Table 3.2: Hillside School District Measure of Aligned Teaching During reading instruction, in your past 5 lessons, how often did you...? During reading instruction, in your past 5 lessons, how often did students...? Read aloud to the class Model fluency and oral reading strategies Model think aloud strategies Model asking and answering questions about texts Model visualization by thinking aloud while reading to students Model strategies for students to learn new vocabulary from texts Ask students to read silently on their own Teach students strategies for decoding sounds and words Individually conference with students while reading Form and work with small groups around reading skills Form and work with small groups around leveled texts Table 3.3: Lakeside School District Measure of Aligned Teachin Read individually Read in small groups Listen to the teacher read Write about what they have read Visualize a text Identify the main ideas of what they have read Compare what they have read with other things that they have read Answer oral questions about or orally summarize what they have read Talk with other students about what they have read 58 Participants were given the following four response choices: All 5 lessons; 3-4 lessons; 1- 2 lessons; or 0 lessons. Consequently, each response was given a 5 for all 5 lessons, a 3 for 3-4 lessons, a 1 for 1-2 lessons and a 0 for 0 lessons. For both of these districts, each teacher’s responses to the survey items in Figure 3.3 or Figure 3.4 were averaged to estimate the teacher’s average use of aligned reading instructional practices over the last 5 lessons. A separate measure was created for the fall and spring survey administrations. Network Exposure. The network exposure term was constructed from teacher’s responses to the survey item “Who do you talk to in your school building about your ELA instruction?” and weighted by the frequency indicated in the follow-up question “How often do you talk to this person?” I used the self-reported measure of aligned teaching from time 1 for each of the teacher’s nominated colleagues to form a measure of their colleagues’ influence on the participant. I labeled this construct network exposure because this term represents an estimate of the information regarding reading instruction that a teacher could have been exposed to in her social network. This construct provided an estimate on the amount of influence a teacher’s colleagues could have had in helping that teacher alter her practices in ELA instruction from time 1 to time 2. Planning Using District Instructional Resources. To create a measure of teachers’ use of district-provided reading instruction resources when planning their reading lessons, I used the question: “I use the following instructional resources / supports when preparing my ELA lessons….” (see Figure 3.5 for the lists of resources in each district). Participants indicated whether they: strongly agreed (4), agreed (3), disagreed (2), or strongly disagreed (1) for each resource. I decided to only include items that were district-provided for their reading 59 instructional policy to measure teachers’ aligned planning. See Table 3.4 for the resources for each district. Hillside School District Lakeside School District I use the following instructional resources / supports when preparing my ELA lessons - Journeys Book ELA Curriculum Framework Making Meaning Book C.A.F.E./ Daily 5 Resources and Website Pensieve Website Table 3.4: Measure of Planning with District Resources For each teacher, these items were averaged to form the teacher’s measure of planning with district resources. Perceived Program Coherence. Drawing off of Newmann et al. (2013), in the spring administration of the survey, I asked questions regarding the participants’ beliefs on program coherence with the questions shown in Table 3.5. These survey items were on four-point scale with the response options of strongly agree (4), agree (3), disagree (2), and strongly disagree (1) scale. Participant responses were averaged to create each teacher’s measure of perceived program coherence. Measure of Program Coherence In this school... - There is consistency in ELA curriculum and learning materials among teachers in the same grade level In this school... - There is consistency in ELA instructional strategies among teachers in the same grade level In this school... - I see real continuity from one program to another In this school... - We have so many different programs that I can't keep track of them all In this school... - Most changes introduced have little relation to teachers' and students' real needs and interests Table 3.5: Measure of Perceived Coherence Professional Fit. Drawing from Organizing Schools for Improvement: Lessons from Chicago (Bryk et al., 2010), I used the following survey items in Table 3.6 to measure teachers’ feeling of connection to their schools. These survey items were included in both the fall and spring administrations of the survey. They were on a four-point scale with the response options of strongly agree (4), agree (3), disagree (2), and strongly disagree (1) scale. Participant responses were averaged to create each teacher’s measure of perceived professional fit. 60 Measure of Professional Fit In this school... - Other teachers matter to me In this school... - I feel like I belong In this school... - I matter to other teachers In this school... - I identify with other teachers In this school... - Most teachers believe that students' well-being is important In this school... - Most of my colleagues share my beliefs and values about teaching Table 3.6: Measure of Perceived Professional Fit Teacher Background Characteristics. Teacher background characteristics were also collected on surveys. These characteristics included a teacher’s total years teaching, her background in reading as defined by a minor or advanced degree in reading or ELA, the grade level she teaches, and gender. Data Analysis All interviews with the superintendents, building leaders, and focus teachers were audio recorded and transcribed. The qualitative data (observation field notes, interview transcripts, and documents) were organized and coded. Throughout my year of data collection, I wrote analytic memos after each network meeting observation to add more context to the field notes taken during the data analysis. These analytic memos were also organized and annotated for easy reference during data analysis. Research Question 1: Theory of Action for the Blueprint for Systemic Reconfiguration In this first research question, I sought to determine the theories of action used in the development of the Blueprint model in general and more specifically the theory of action for influencing instruction at scale. In order to construct the MI Excel “Blueprint for Systemic Reconfiguration,” I drew from MI Excel online resources and my professional development experiences, such as the “Diving Deeper into the Blueprint” level I, II, and III trainings that I attended over the 2016-2017 school years. These documents had some limitations in that they primarily attempted to inform readers about what each component was and how it would look in 61 a district, without addressing the logic behind why that component was included in the Blueprint framework and explicitly how it would contribute to improved teachers’ practices and student learning. Thus, to distill the theory of action of the Blueprint in general, I had to synthesize multiple documents created by MI Excel. I began with their document entitled The Blueprint’s Theory of Action by Chandler and Frank (2017). I then coded additional online resources from the MI Excel online database and training documents for how it fit into the theory of action presented in this document. A limitation of the original theory of action provided from MI Excel leaders (see Figure 3.2) was that it presents the Blueprint process as a circle that yields a “dramatic increase in student, teacher, and leader performance.” However, each system was described as having three components that build on each other in a more linear manner and the implementation timeline (see Appendix A) provided a more straightforward presentation of how systems build on one another to form the Blueprint, which in turn would influence student, teacher, and leader performance. 62 Figure 3.2: Blueprint for Systemic Reconfiguration Theory of Action (Chandler & Frank, 2017) In terms of distilling the theory of action in regard to changing instruction explicitly, I referred to documents related to the instructional infrastructure system of the Blueprint. More specifically, I looked for common terms in the documents to form a logic map of how the components fit together sequentially to lead to the potential of improved teacher instruction and student outcomes. In creating this theory of action model, I relied on the documents from the MI Excel Online Resource Center entitled Instructional Infrastructure Evidence of Practice, Instructional Infrastructure Training PowerPoint, and the Curriculum and Assessment Training PowerPoint. 63 Research Question 2: Administrator Orientation to Change and District Theory of Action In this second research question, I wanted to build on the MI Excel version of the theory of action for the Blueprint to identify how district and school leaders’ assumptions regarding changing instruction at scale influenced their district-specific theory of action. I used network observations, district documents regarding the Blueprint, and interviews with superintendents, building leaders, and teachers to respond to this question. First, in working to define district and school leaders’ orientation to change, I analyzed the qualitative data listed using an inductive coding strategy to develop codes related to administrators’ beliefs around influencing change and their beliefs about the Blueprint’s potential to create change. Inductive coding refers to the practice of allowing codes to emerge from the data during data collection and analysis. Inductive coding ensures that the coding was sensitive to important local factors (Miles, Huberman, & Saldaña, 2014). When coding, I began with a descriptive coding process to identify teacher and leader characteristics (Miles et al., 2014). I then grouped codes into categories examining causes/explanations, relationships among people, and theoretical constructs. Miles and colleagues (2014) refer to this process as pattern coding. Finally, I completed a round of selective coding in which I re-coded the transcripts into the Lewis (2015) framework comparing the experimental science paradigm and the improvement science paradigm (see Figure 2.1). I found this framework to validly reflect the assumptions shared by each district and school leader. As I recoded the transcripts, I was also sensitive to disconfirming evidence that would indicate that this framework did not match the administrators’ orientations toward change (Strauss & Corbin, 1998). I did not find any codes referring to an administrator’s beliefs about the change process that did not fit into the Lewis (2015) framework. 64 In order to construct each district’s theory of change for impacting instruction at scale, I drew on network observations, interviews with district and school leaders, and document analysis. I inductively coded these forms of evidence to create as many categories as possible related to the district’s plan to influence teacher instructional practices. From these inductive codes, I engaged in pattern codes in which I grouped the initial codes into a smaller number of categories exploring causes/explanations, relationships among people, and theoretical constructs (Miles et al., 2014). Some categories that emerged include “the BTN as instructional coaches” and “high-quality instruction walkthroughs.” Finally, I examined the categories created from the pattern codes to build a logic model (Miles et al., 2014; Yin, 2014). This logic model served as a diagram to sort out relationships among concepts in each district Logic models operationalize chains of events over an extended period of time (Yin, 2014). Additionally, creating these logic models provided a reference when completing data analysis for research question 2 and in identifying adaptions to the original MI Excel theory of action in each district. Brickmayer and Weiss (2000) contend that this type of analysis can make visible the theories and assumptions underlying a program as well as allow a researcher to follow each step to see whether the expected mini-steps actually occur. Research Question 3: Implementation of Each District’s Theory of Action The focus of the third research question was on how the previously described leadership orientations towards change impacted their implementation of their theory of action into their schools. To answer this question, I used network observations, district documents regarding the Blueprint, classroom observations, and interviews with superintendents, building leaders, and teachers. Like with the previous research question, I began by inductively coded in which I developed codes related to the reading program and Blueprint implementation (Miles et al., 65 2014). I then categorized the codes examining causes/explanations, relationships among people, and theoretical constructs in a cycle of coding referred to as pattern coding (Miles et al., 2014). These categories included the constructs of “mandated programs and activities,” “training activities,” and “monitoring”. Finally, I completed a third round of coding referred to as selective coding in which I used these categories to create a network display for each district in which I mapped how the leadership assumptions from the previous research question connected to the categories that emerged during pattern coding (Miles et al., 2014; Strauss & Corbin, 1998). This network display allowed me to visually compare the differences between district implementation of the instructional infrastructure into their schools (Miles et al., 2014). Research Question 4: Changes in Instructional Practice and Planning The focus of the fourth research question was on how teachers change their instructional practice and planning in response to the Blueprint work. To this end, I analyzed the classroom observation data in relation to the ELA subject-specific framework developed by each district (see Appendix B for the example created by MI Excel). I deductively coded the observation field notes using each district’s ELA subject-specific framework as a guide (Miles et al., 2014). This coding allowed me to collect examples of teacher and student actions under each recommended practice. These examples were then compared for quality and depth in order to sort teachers into implementation categories. This coding also allowed for comparisons between observations from the same teacher across the school year as well as comparisons between teachers. I also considered the focus teachers’ interviews regarding planning in responding to this question. Again, I began by inductively coding for patterns around teacher planning processes, beliefs about the reading instruction reforms, and beliefs about the Blueprint reforms overall (Miles et al., 2014). I then grouped the codes into categories. From these categories, I compared 66 teachers using a matrix display that allowed me to compare individual teachers’ responses over time (Miles et al., 2014). This matrix display also allowed for a cross-case synthesis in which I could compare codes from teachers at different schools and districts to look for differences between cases (Yin, 2014). To represent the structure of the advice network of each focus school and the location of each focus teacher, I applied Kliquefinder (Frank, 1995) to the collegial ties that participants indicated in the survey question, “Since the beginning of the school year, with whom in your school building do you discuss your ELA instruction?” Kliquefinder indicates the structure of subgroups within a social structure, such as a school. Consequently, I utilized Kliquefinder to uncover the structure of subgroups separately in Hickory Elementary School, Willow Elementary School, Cottonwood Elementary School, and Roe Elementary School. Quantitatively, I used the previously described surveys completed by all teachers in the focus schools to estimate the factors that may influence a teacher to change their instruction from Fall 2016 to Spring 2017 in a way that was aligned with district expectations. The survey asked questions related to a teacher’s reading instructional practices, planning practices, collaboration opportunities, and beliefs regarding the reforms. The survey included data collected at two different times over the school year. This data was used to complete the following influence model in order to see what components significantly influenced teachers to change their reading teaching practices: Outcome = aligned teaching practices at time 2 Independent variables of interest = network exposure, planning using district instructional resources, perceived program coherence, professional fit i = teacher i'= nominated teacher t = time 67 ("#$%&'( *'+,-./,&0'$1 2-$/,&/#+ $, ,&3# 2)67 = 9:+ <∑6>?@A (B#1C)66>7D@→7(B#1C C-0F&%#-+G#+,&3$,#% C-$/,&/#+ $, ,&3# 1)6>7D@ + 9@ (Reading Instructional practices at time 1)+9Z (Missing Exposure Term Flag) + 9` (Planning using District Resources) +9c (Beliefs that the Policies are Coherent)+ +9h (Feelings of Fit in School) + 9j (District Fixed Effects)+γ (Vector of Teacher Characteristics)+#6 Equation 3.1: OLS Model of Factors Related to Aligned Teaching in Spring 2017 Overall this model allowed me to estimate the influence that a teacher’s colleagues, her planning practices, her beliefs about district policies, and her feelings of professional fit have on her change in reading teaching practices during the 2016-2017 school year. Summary of Methods In sum, my data collection and data analysis methods answered my research questions using the following approaches: How, if at all, does an improvement science-based reform influence the instructional core at Overarching Question scale throughout a school district? Data Analysis Overview Data Analysis –inductive coded (Miles, Huberman, & Saldaña, 2014) and logic models (Yin, 2014). Sub-Research Question a) What is the intended theory of change of the Blueprint for Systemic Reconfiguration for influencing instruction at scale? Data Collection Overview Observation – Took field notes around Blueprint systems and drivers at MI Excel trainings and institute days created for the professional learning of leaders in districts implementing the Blueprint. Document Collection – Collected documents and copies of PowerPoints. Downloaded resources and informational material from the MI Excel Resource Center online. Table 3.7: Summary of Methods 68 Data Analysis –inductive coding (Miles, Huberman, & Saldaña, 2014) and logic models (Yin, 2014). Data Analysis – inductively coded (Miles et al., 2014) and cross-case synthesis (Yin, 2014). Table 3.7 (cont’d) b) How does each administrator’s orientation toward change influence the implementation of Blueprint for Systemic Reconfiguration in their district or school? c) How is the theory of change enacted in each district? How does this enactment reflect administrators’ orientations toward change? Observation – Took field notes on the topics covered at district network meetings and building meetings. Interviews - Interviewed superintendent, principal, and a sample of teachers to understand their goals and theory of change Document Collection – Collected agendas and documents used at network meetings. Downloaded information from district websites on turnaround. Observation – Took field notes on the topics covered at district network meetings and building network meetings. Also observed and took field notes at PSP meetings and a sample of staff meetings. Interviews - Interviewed superintendent, principal, and sample of teachers to understand their implementation plans. Document Collection – Collected agendas and documents used at network meetings. Downloaded information from district websites on turnaround. 69 Table 3.7 (cont’d) d) To what degree and in what ways does the enactment of the theory of change influence teacher instructional practices and planning in each school? Observations – Compared field notes compared to district frameworks using deductive coding (Miles et al., 2014) Interview – inductive coding (Miles et al., 2014) and cross- case synthesis (Yin, 2014). Survey - Social Network Influence Model Class Observations of Focus Teachers – Observed and took field notes of teacher reading instruction. Interview – interviewed focus teachers about the changes they were working on making in their classroom and their planning processes Survey - Surveyed all teachers in the schools about their ELA instructional strategies and their social networks: “Who do you talk to in your school building/ district about your ELA instruction?” Limitations As with all research, this dissertation does have some limitations. First, this study was limited to participants in the two districts. While originally chosen for their similarities, these two districts ended up representing different assumptions around change processes and leadership. This resulted in each district representing a different change paradigm. This leaves the possibility that a different district implementing the Blueprint under the experimental paradigm or the improvement science paradigm could have gotten different responses and implementation levels from their teachers than were found at Hillside and Lakeside. Moreover, these districts were not randomly selected, but were recommended by the director of the Blueprint program as districts who were early adopters and implementing the Blueprint for Systemic Reconfiguration a high level. Furthermore, the focus elementary schools and teachers were also not randomly selected within districts. District superintendents and school 70 principals chose which elementary schools and classrooms they thought would be best for me to observe given my criteria of maximizing variation in the leadership levels of each teacher. Given this limitation, it was interesting that in the Hillside School District I observed two teachers who were resisters to the reforms. With this limitation in mind, I also intentionally ate lunch in the teacher’s lounge of each school when completing an observation so that I could hear the opinions of most teachers. This practice gave me access to the opinions of teachers not participating as focus teachers in this study in order to listen for differences between the opinions and experiences of other teachers at the school in comparison with those teachers participating in this study. The information that I heard in the schools’ teacher lounge in reference to the reading curriculum and Blueprint were largely consistent with the opinions provided by the focal teachers in their debriefing interviews. Lastly, the use of surveys shared with all teachers in each of my four focus elementary schools served as another way to gain access to the opinions and experiences of all teachers, not just the 12 observed and interviewed teachers. The second limitation is that while I was observing the district leaders, school leaders, and teachers in their natural environment as they worked to implement the systems of the Blueprint for Systemic Reconfiguration and the accompanying district literacy expectations, my presence as a researcher could have influenced their actions. I attempted to overcome this limitation by observing DN and BN multiple times over the course of the school year and teachers’ instructional strategies three or four times from October 2016 to June 2017. Hopefully, these multiple observations allowed participants to feel more comfortable with my presence and interact how they would regularly without an observer. Despite these limitations, the combination of multiple data sources in the form of documents, observations, interviews, and surveys provides insight into how an improvement 71 science program impacts practice at all three levels of a school organization: district, school, and classroom. As policymakers and practitioners continue to grapple with how to improve student and teacher performance, the results and analysis of this dissertation provides hypotheses to consider around leadership assumptions about change and teacher reactions to changes under an improvement science paradigm. Establishing Validity Yin (2014) outlines four tests of validity for case study research: construct validity, internal validity, external validity, and reliability. First, research must have construct validity meaning that this study identifies “correct operational measures for concepts being studied” (p. 46). In order to meet this test of validity, a researcher must define specific research concepts and identify measures that match the concepts (Yin, 2014). In this study, I define my research concepts of “impacting teacher instructional practices” as teachers teaching reading in a manner aligned with district policies. I define this concept using documents created by each district outlining expectations for high-quality reading instruction, such as the district visions of high- quality instruction and implementation guides. I then examine this construct using multiple sources of evidence including field notes from district network and building network meetings, field notes from classroom observations, teacher interviews regarding their planning practice and instructional decisions, and lastly through survey measures. The second test, internal validity, describes the truth value of qualitative research (Miles et al., 2014). Miles and colleagues (2014) define this test as “the account rings true, makes sense, seems convincing or plausible, and enables a vicarious presence for the reader” (p. 313). In this study, I respond to this test through inductive coding with pattern building, by addressing rival 72 explanations, considering negative evidence, and through using logic models (Miles et al., 2014; Yin, 2014). The third test, external validity, concerns the domain to which the study’s findings can be generalized (Yin, 2014). This test can be fulfilled through asking “how” and “why” questions that require analytic generalization (Yin, 2014). In this study, I consider three “how” questions: how does each administrators’ orientation toward change influence the implementation of the Blueprint for Systemic Reconfiguration in their district and school?; how is the theory of change enacted in each district?; and how does the enactment of the theory of change impact teacher instructional practices? Additionally, this case study strived to meet this test by choosing two districts using replication logic, in which both districts were predicted to have similar results when implementing the Blueprint for Strategic Reconfiguration and its influence on teachers’ instructional strategies (Yin, 2014). Moreover, the sampling logic of teachers within each district allowed for adequate comparisons based on involvement in leadership positions as well as replication between teachers in each district (Miles et al., 2014; Yin, 2014). This combination of sampling for adequate comparison as well as allowing for some replication provides evidence for external validity of the results (Miles et al., 2014; Yin, 2014). The last test, reliability, ensures that if a future researcher follows the same procedures and conducts the case study again, the “later researcher should arrive at the same findings and conclusions” (Yin, 2014, p. 48). This test can be met through establishing clear research questions; establishing and describing the researchers role; collecting data over a range of settings, times, and participants; and connecting analytic constructs to theories (Miles et al., 2014). As described in greater detail above, my research design addressed the test of reliability through having clear research questions, collecting data from 18 participants over the course of 73 seven months, collecting both qualitative and quantitative data, and by connecting research constructs to the experimental and improvement science paradigms. Overall, this study’s data collection and analysis procedures met Yin’s (2014) four tests of validity for case study research. 74 CHAPTER 4: THEORY OF ACTION AND LEADERSHIP ORIENTATIONS TO CHANGE This chapter answers my first and second research questions: (1) What is the intended In my two case study districts, Hillside and Lakeside, district superintendents took theory of change of the Blueprint for Systemic Reconfiguration for influencing the instructional core? and (2) How does each administrator’s orientation toward change influence the implementation of Blueprint for Systemic Reconfiguration in their district or school? Overall, the Blueprint for Systemic Reconfiguration, also referred to as the Blueprint, is based on the assumptions of improvement science in that it provides a framework for district leaders to design and roll out systems in their districts that address human resources decisions, instructional expectations, meeting student nonacademic needs, school data analysis protocols, and leadership expectations. District leaders work within guidelines to design systems that they feel will work best in their local environment and with their local expertise. significantly different paths in working with the Blueprint. In this chapter, I will summarize each administrator’s orientation toward change and their actions under the district turnaround program, the Blueprint for Systemic Reconfiguration. I then will outline how the Hillside superintendent and one of her principals applied experimental science assumptions to the Blueprint in that they looked to implement the Blueprint and the district designed systems with fidelity and worked to monitor teachers’ implementation fidelity. In contrast, at Lakeside the superintendent and principals applied improvement science assumptions to the Blueprint as the district refined and modified Blueprint information to best meet the needs in their district. Additionally, school principals regularly modified district provided resources in their work in their schools and shared their revisions with others in the district. 75 Theory of Action for the Blueprint for Systemic Reconfiguration The Blueprint for Systemic Reconfiguration focused on the district level and called on central office staff to lead the effort to improve instruction around district schools and then to create systems that support that instructional improvement (Chandler, Mohney, & Brophy, 2017). Creators of the Blueprint for Systemic Reconfiguration define the Blueprint as follows: The Blueprint for Systemic Reconfiguration is grounded in the research of experts in reform research but takes their work on specific elements [created in isolation] and connects them to each other to create a powerful, systemic approach that will challenge the status quo, base the organization’s decisions on what is needed to support student learning for all students, and provide the support and guidance needed to realize and sustain dramatic improvement. (Chandler & Frank, 2015, Slide 77) The creators of the Blueprint for Systemic Reconfiguration argued that the power in the Blueprint was that it brought together siloed research around school improvement into one program. During summer institute for districts interested in the Blueprint, the presenter stated, “the Blueprint provides a strategic roadmap for districts to build and install systems that are most closely aligned to student achievement.” (Hoppa, 2017). The program led district leaders through completing a series of planning tools to coordinate the design and implementation of a variety of systems addressing such organizational functions such as (1) human resources, (2) curriculum and instruction, and (3) student services. The Blueprint for Systemic Reconfiguration rested on the assumption that the current systems and routines in place in a district were inadequate to meet the academic and nonacademic needs of students (Chandler, 2017). Blueprint leaders claimed that low performing districts would improve student academic performance if they do the following actions: 76 (1) Establish urgency for the success of all students (2) Install key coherent systems (talent management, instructional infrastructure, leadership network, and intense support network) (3) Develop and support impactful instructional leaders who skillfully leverage these district systems to support exemplary instruction of rigorous standards-based curriculum and intensive support of student’s non-academic needs (4) Install highly effective driver systems (problem-solving, communications, and performance management) to realize district configuration at scale. If districts completed these four actions, then the Blueprint asserted that collective responsibility would occur among school leadership teams and that, through more aligned district support both student and teacher performance would improve in a short amount of time (Chandler, 2017). The Blueprint facilitated these four actions through the installation of systems at both the district and school levels (See Appendix N for the turnaround logic model). First, leaders were to establish a district network (DN). This DN typically was comprised of a school board member, superintendent, and central office administrators. The creators of the Blueprint stressed that the superintendent should be the one to lead this work, because he or she could tap into all of the district departments to coordinate the turnaround work. Additionally, having the superintendent lead the effort served as a signal to others in the organization of the seriousness of this work and that it was a district priority. Throughout the three years of Blueprint implementation (see Appendix A for the Blueprint Installation Timeline), the DN met weekly. The creators of MI Excel stressed that the DN should have norms around respecting the confidentiality of items discussed in the room, allowing for all voices and perspectives to be heard before any decision was made, and 77 supporting the DN’s decisions in members’ roles around the organization. The DN had three main purposes. First it led the district turnaround efforts. Second, it worked to create and sustain a sense of urgency for change around the district. Third, it monitored implemented systems to ensure that they all are implemented to scale (English & Robinson, 2017). A significant part of DN meetings included discussing and filling out planning tools on each component of the systems that they planned to implement throughout their districts. These planning tools required DN members to write out different aspects about the purpose of the system, the processes within the system, communication plan, implementation process, and modification processes. (see sample instructional infrastructure planning tool in Appendix O). At the district level, the DN primarily focused on installing three systems: instructional infrastructure, talent management, and intense student support network. This study primarily focuses on the instructional infrastructure system, so that will be discussed in more detail below. The second system, talent management, created a new structure concerned with finding, identifying, and retaining the talent in the district necessary to influence student learning and teacher performance. It focused on establishing and communicating clear expectations for staff districtwide in terms of “turnaround competencies.” This system had three components: (1) recruitment, hiring, and assignment, (2) evaluation, and (3) retention and removal. Under the first component, the DN worked to create a list of turnaround competencies that they expect in every employee in the district and created procedures for assigning personnel and for recruiting new employees. Under the second component, evaluation, DN members examined their evaluation instruments, created data systems to inform, support, and align other policies and procedures to the evaluation system. In the third component, retention and removal, DN members defined their retention and removal procedures and refined their data systems. 78 The third district system was the “Intense Student Support Network.” This system focused on student nonacademic needs. The assumptions behind this system was that schools needed to provide pastoral care for their students (Murphy & Torre, 2014). Again, there were three subcomponents within this system: (1) Vision of High-Quality Student Support, (2) Identifying and Meeting the Social, Emotional, Health, and Nutritional Needs of All Students, and (3) System of Network Delivery. In the first subcomponent, the DN worked to develop a vision document and share it with all members of the district. In the second subcomponent, the DN began to collect data on student needs, develop policies that address those needs, and train staff on meeting student nonacademic needs. In the third subcomponent, data is collected on the system, district policies were developed or modified to influence adult and student interactions, and the district found opportunities to foster student voice. The DN was also responsible for building the three driver systems that connected the work at the district level to the work that happened at the school level. These drivers were: the communications driver, the problem-solving driver, and the performance management driver (English & Robinson, 2017). The communications driver provided systems for how information would be communicated around the district. More specifically, this system planned for how information related to the Blueprint would be communicated to stakeholders. It also designed a plan for how information from schools and teachers could be communicated to district leaders. The assumptions behind this system were that (a) influencing change required all organizational members to be familiar with the vision, goals, and models of the reform, and (b) leaders of organizational change required feedback in order to effectively lead (MI Excel Statewide Field Team, 2017a). The second driver system was the problem-solving driver (commonly known as the PSP). This driver called on district leaders to design a process for facilitating effective data 79 conversations. The assumption behind this system was that data literacy and use could shift the focus of a reform policy from compliance to commitment through a shared goal. This system required district leaders to develop a process for data collection and opportunities for data conversations using multiple measures of data that develop meaningful action steps to inform future behavior (Brockway, 2017). The last driver system concerned monitoring the systems and building implementation. The Performance Management (PMT) driver enabled district leaders to understand the quality of Blueprint implementation in terms of (1) the extent to which systems had been installed at scale throughout the district and (2) the extent to which building leaders and their staff were analyzing multiple measures of data to understand their progress meeting their goals. The creators of the Blueprint recommended that this system occur in the form of a monthly meeting in which building leaders present their data and building progress to district leaders and other school leaders. In these meetings, conversations between leaders should include questions about building processes and challenges and collectively the group could engage in problem-solving (MI Excel Statewide Field Team, 2017c). While the district network was beginning its conversations and planning, school leaders needed to establish building networks (BNs). The creators of the Blueprint recommended that BNs begin after four months into installation of the Blueprint and the BN be comprised of the building principal, any other building administrators, and teacher leaders. The role of the BN was to navigate and lead the implementation of the district developed systems into their school environment. Additionally, at BN meetings, the group should be reviewing building data prior to performance management meetings so that the principal has a stronger understanding of the potential reasons behind the data (English & Robinson, 2017). Moreover, the building network was responsible for leading three systems at the building level: (1) Principals Leveraging District 80 Systems, (2) Instructional Leadership Routines, and (3) Teacher Collaborative Routines. The first system was concerned with the principal leveraging district systems and drivers to support systemic reconfiguration in his or her building. There were three subcomponents to this system. First principals were to know how to communicate regarding their student academic and nonacademic data and be able to communicate their needs. Second, principals were to utilize district systems such as talent management, leadership network, instructional infrastructure, and intense student support network to meet the specific needs of the students in their care. Finally, principals developed a partnership with one member of central office to learn and better utilize district systems and supports. The second building-level network was Instructional Leadership. This system was designed to support principals and other building leaders to become instructional leaders that increase teacher and student performance. Again, this system had three subcomponents. First, principals were to promote alignment between the district’s vision for subject-specific, high- quality teaching and the instructional practices being used in their classrooms. This included providing opportunities for teachers to collaborate around the materials and developing learning plans around the vision. At the second level, principals were to act as coaches for their teachers by communicating the district’s vision, regularly observing instruction, providing feedback on instructional practices, modeling instruction consistent with the vision, and identifying further supports for teachers struggling to teach aligned with the vision. Under the last subcomponent, principals were to continue to monitor the instructional environment through continued classroom observations, monitoring curriculum planning and implementation, collecting data on teacher and student performance, and examining student work. 81 The last building level network was Teacher Collaborative Routines. This was one of the last systems implemented under the Blueprint Installation Framework. This system was designed to provide teachers opportunities to support and guide each other as they worked to improve their instructional practices. This system also had three subcomponents. First, teachers collaborated to match their teaching to each other through jointly planning and matching their instruction to the curriculum and vision. Second, teachers met to design and analyze formative data on student learning and make any necessary adjustments. Third, teachers had opportunities to observe each other teaching, provide informal feedback, and model lessons for one another. Collectively all of these systems, drivers, and routines implemented at both the district and school levels should work together to disrupt current district, school, and teacher routines and substitute a coherent structure that supports student success (Frank & Chandler, 2016). Overall, this structure of systems focused on supporting high-quality teaching and deep student learning. In the next section, I build on this background of the systems to further outline the theory of action for how these systems will influence each teacher’s instructional practices. Theory of Action for the Instructional Infrastructure More specifically, the Blueprint centered around the notion that “turnaround requires instructional improvement at scale.” Therefore, in this study, I focused on the instructional infrastructure system to explore how district systems influence classroom instruction. The creators of the Blueprint defined this system as “concerned with implementing and supporting a vision of high-quality, ambitious subject-specific instruction to impact student/teacher performance in districts where systemic reconfiguration purposefully disrupts current practice to create a new structure of coherent and aligned district and building systems to ensure success for all students.” (Irving & Mohney, 2017, slide 12). 82 Figure 4.1: Theory of Action for Instructional Infrastructure 83 Drawing from training materials and other documents created to assist districts in implementing an instructional infrastructure, I created Figure 4.1 to show the theory of action for influencing teaching practices at scale throughout a district. The green squares describe the steps in influencing teaching practices, while the purple ovals describe the district actions within each step. First, district leaders created a vision of high-quality instruction (HQI) for each subject area and communicate it to stakeholders around the district in order to develop a shared understanding. Appendix C shows the example vision of high quality instruction for reading that MI Excel offered as a model for districts. After creating subject specific visions for HQI, district leaders must find ways to communicate this vision to staff. MI Excel recommended that this happen through publishing vision documents in some way that made them accessible for district employees, such as on a website (MI Excel Statewide Field Team, 2017b). Training materials also included a template for teachers to work through at a staff meeting in order to work collaboratively to become familiar with the district’s vision of HQI (Liberato, 2017). It was expected that over time, these visions would influence human resources decisions, curriculum and assessments, teacher evaluation processes, and professional development throughout the district (Liberato, 2017). In the second step, district leaders and other committees worked to design curriculum that is aligned to the vision for HQI. This curriculum included a curriculum framework that specified learning targets, sequence, pace, and units of instruction for each subject area and grade level. These frameworks were to also include nonacademic targets, such as learning goals for social, emotional, health, and nutritional learning. In these frameworks, MI Excel endorsed an “Understanding by Design” format (Wiggins, McTighe, Burz, & Wood, 2011) in which district leaders began by defining the goals for student learning at the end of the unit and then planned 84 the components accordingly. Curriculum also included curricular materials aligned to the vision of HQI. Assessments aligned with the HQI visions and curriculum frameworks were the last part of the curriculum. These assessments also had to be sensitive enough to identify students who may need additional support. Districts would complete this step when they published their curriculum framework and assessments for all staff to review and use. Additionally, in this step, district leaders were to develop a protocol for central office and school leaders to complete high quality walkthroughs (HQI walkthroughs) that would allow these leaders to get an idea of staff’s alignment of instructional practices to the vision of HQI and their use of supporting curricular materials (See Lakeside School District’s observation form for HQ walkthroughs in Appendix P) (MI Excel Statewide Field Team, 2017b). After curriculum and assessments were created, the DN was to define some sort of approval process for the documents prior to their release to staff. The model planning tool (MI Excel Statewide Field Team, 2017b), described a process in which teachers and administrators would develop any needed curriculum documents and assessments. Following their completion, a district designee would review, provide feedback, lead the modification process, and provide preliminary approval. Finally, the superintendent and school board would decide on final district approval. This document also included plans for annual review of curriculum maps in June of each year. This plan included a process for district leaders to review feedback from teachers; make revision requests of the curriculum developers; a process for re-approving the modified curriculum; and a plan to roll out the revisions. Once curricular materials and assessments aligned to the district vision of high quality instruction were approved, it was expected that teachers begin teaching in alignment to the expectations of these frameworks. In the Differentiated Instructional Improvement Network 85 component, district and school leaders began to collect and use data on teacher and student performance to identify the needs in the district. In this step, district and school leaders collected data on inventory lists of curricular materials available, evidence of the frequency and time in which instructional leaders and teachers were examining data, number of students identified for instructional intervention, number of students at each intensity level of instructional supports beyond tier one, evidence of impact of new curriculum system on student learning, and evidence of instructional practices in the classroom from teacher observations (MI Excel Statewide Field Team, 2017b). This step also related to support for teachers as they worked to adjust instructional practices in alignment with the visions for high quality instruction and the new curricular frameworks. District leaders provided differentiated professional learning for teachers aligned to the visions for high quality instruction that supported teachers’ understanding of curricular frameworks, materials, and assessments. Additionally, the district was to designate and protect time for teachers to collaborate around the visions for HQI and curricular frameworks, materials, and assessments. Evidence of this occurring would be in the number of hours dedicated for teacher collaboration and the number of teachers involved. This teacher collaboration time also was to include time for teachers to collaborate around improving student performance through examining student data and work. Lastly, district leaders were to provide coaches or other specialists who can support teacher learning and collaboration around improving their instructional practices. Evidence of this occurring would be evident in the list of coaches or content specialists available to teachers as well as the frequency of coach and teacher interactions and the focus of these interactions. Much of this step in the process also connected to the building level instructional leadership and teacher collaborative routines systems described above. 86 Collectively these steps should increase adult capacity at scale across a district to teach in alignment to a vision of high-quality instruction. Under the belief that increasing teacher competencies would lead to teacher actions, which would lead to improved outcomes (see Figure 4.2). Figure 4.2: Theory of Action for Staff Development in Instructional Infrastructure MI Excel staff designed a system for instructional infrastructure in which district leaders defined and shared with staff goals for instructional practices and curriculum. They then monitored teacher actions through high-quality walkthroughs and monitored outcomes from data on those high-quality walkthroughs and student achievement. 87 Blueprint and Improvement Science Overall, the Blueprint for Systemic Reconfiguration was based on improvement science assumptions in that the Blueprint was a framework that allows local actors to use their knowledge of local systems in designing and implementing change in their districts. Blueprint leaders summarized the action of implementing the Blueprint in the following way, “The word install signifies something permanent, something that will need to be monitored and adjusted, something connected to a bigger system, and something interdependent on other systems” (Hoppa, 2017, slide 14). This quote highlighted that even while administrators implementing the Blueprint were working towards addressing systems in a lasting way in their districts, they should also expect modifications over time. This quote also reinforced that overall the Blueprint addressed organizational systems and that these interdependent systems determined the success of organizational work (Perla et al., 2013). District leaders on the DN met regularly using the framework provided by the Blueprint of topics and district systems to discuss. These DN meetings provided opportunities, like in NICs (Bryk et al., 2011), for stakeholders to exchange their knowledge and to collectively design changes. By design, DNs were to be collaborative and open spaces where each member had the ability to add a topic to the agenda, share their experiences, and collectively design solutions. Planning tools provided by Blueprint leaders guide the discussions in DNs to maintain a systems- focus. Similar to DNs, BNs met regularly to share knowledge regarding how implementation of the district-designed systems was going. These meetings also worked like NICs in that members from diverse perspectives around the school were able to discuss any problems, develop possible interventions, and lead the use of these interventions throughout the school. For BNs, DNs are 88 able to act as their coordinating hubs for the sharing of common problems and the sharing of improvement ideas among BNs. Beyond the network meetings providing a space for stakeholders to share their knowledge, the Blueprint also organized PDSA cycles (Bryk et al., 2011; Perla et al., 2013) in PSP and PMT meetings. As described above, PSP meetings happened at the school level and gave staff opportunities to look at their current student data, make a plan to address a deficiency in the data (plan), attempt their plan in their classroom over the course of a month (do), analyze new student achievement data from that month (study), and then decide how to adjust their instructional plan (act). PMT meetings worked similarly to PSPs but occur among district leadership and school administration. Collectively, the Blueprint for Systemic Reconfiguration represented a framework for incorporating improvement science assumptions about change into the work of district and school improvement. It relied primarily on systems reforms and organized NIC-like groups for defining problems and designing solutions. These solutions and system-reforms were assessed through PDSA cycles at the building and district levels in PSP and PMT meetings. Theories of Action in Practice Both Hillside School District and Lakeside School District worked to implement the Blueprint for Systemic Reconfiguration into their schools during the 2016-2017 school year. In both districts, this involved weekly DN meetings to complete planning tools and discuss current systems and those in development, monthly performance management (PMT) meetings to get an idea of how implementation was going in each school site, and regular BN meetings in each school. Despite these similarities in the structures and routines occurring in each district, there were stark differences in the assumptions behind implementing change between superintendents. 89 These differences influenced the design of the respective district systems, the actions of their principals in leading implementation at their schools, and eventually teacher instruction and beliefs about the Blueprint. Superintendent’s Role MI Excel stressed the importance of the superintendent in district reconfiguration. In fact, before beginning the Blueprint, interested superintendents were interviewed by MI Excel staff to assess their district’s readiness to begin the process. The superintendent interview covered topics around their rationale for wanting to begin the Blueprint, their willingness to change their own professional practices, and their readiness to serve as a turnaround leader (Chandler, 2017). The creators of the Blueprint often quoted the following excerpt from Heather Zavadsky (2012) to argue for the central role of the superintendent in leading the Blueprint: Placing oversight at the highest administrative level– the superintendent’s office –allows leaders of the work to tap into all the above departments and connect the work to the overall strategic plan, and it provides a signal to the district and community that turnaround . . . is a district priority. (p. 213) Following this logic, I began by analyzing the viewpoints of the superintendents. In both Hillside and Lakeside school districts, the superintendent played central roles in defining what the Blueprint was in their districts and the expectations. Their leadership and decisions had great influence on principal actions in their schools and teacher reactions to the Blueprint systems. From observations of DN meetings as well as open-coding administrator interviews, it was very clear that the Hillside superintendent and Lakeside superintendent had very different perspectives on how to improve instruction at their schools and how to implement their plans. I argue that these two superintendents represent assumptions from different paradigms on how to 90 scale up programs. As discussed in Chapter 2, the experimental science paradigm relies on implementing with fidelity any new programs and reducing variation between sites. In the case of the Blueprint, this would involve schools and classrooms implementing with fidelity the programs and routines designed by the DN with minimal variation. In contrast, the improvement science paradigm allows programs to be integrated with local knowledge and expertise. It allows for modifications in programs as data indicates a need. In the case of the Blueprint, this would mean that schools and classrooms may use the systems designed by the DN differently and modifications may be made to DN systems as staff members begin to have experience with the systems. Drawing off of Figure 1 in Lewis (2015) (see Figure 2.1), I deductively coded each administrator’s interview for evidence of beliefs or actions that aligned to either the experimental science paradigm or the improvement science paradigm. Overall, the administrators demonstrated different implementation assumptions with Hillside administrators showing more experimental science assumptions and Lakeside administrators showing more improvement science assumptions. The one exception was the principal at Willow Elementary School in the Hillside School District. I will discuss more about her case below. Table 4.1 shows the number of excepts each administrator had that fell in each paradigm. Number of Excerpts demonstrating an Experimental Science Viewpoint 16 (84%) Number of Excerpts demonstrating an Improvement Science Viewpoint 3 (16%) 19 (83%) 4 (31%) 4 (17%) 9 (69%) Ms. King Hillside Superintendent Ms. Davis Hickory Principal Ms. Nelson Willow Principal Table 4.1: Summary of Administrator Interview Excerpts that Aligned with Either Experimental Science or Improvement Science Viewpoints 91 Table 4.1 (cont’d) Mr. Dixon Lakeside Superintendent Ms. Wilson Cottonwood Principal Ms. Reed Rose Principal 4 (12.5%) 3 (7%) 1 (7%) 28 (87.5%) 40 (93%) 14 (93%) Ms. King – “We adhere to it with fidelity” Ms. King, the superintendent at Hillside entered administration after teaching special education for four years. Before becoming superintendent at Hillside, she worked as a superintendent at another school district in the Midwest. She described herself as a believer in structures and systems: And I really attribute that learning from [person’s name] who was a marine. So you know how Marines really believe in systems and structures, that’s how [former school district] worked that groups reported to groups. …You had cabinet structures. …Everyone had reporting structures and intertwined and planned together, and so it was very systematic how we worked. And I have built my career around that. From this quote, it seemed Ms. King thought of structures and systems in conjunction with the Marines. When thinking about systems and structures in the military, it was clear that she thought about systems in a hierarchical way, which was different way than how the improvement science paradigm would. Military systems are rigid and followed to the letter, while improvement science systems allow for consensus building and sense-making. When discussing her decision to adopt the Blueprint, Ms. King mentioned knowing immediately that this program was what Hillside needed, because in her two years leading the 92 district, she noticed that it, “didn’t have any structure in place and I really believe in systematic structure.” In fact, she signed up during that first training for the Blueprint. Overall Ms. King believed that districtwide change occurred primarily through direction from district leaders. She described district change as occurring as follows: The central office really kind of drives the situation and the curriculum office develops the curriculum and then the buildings, principals their job is to support the teachers in the building and they go in the classroom and that from what they need, we bring supports in and teach that. That system of support is the kind of system of support that was developed and should be developed in large school districts and that everybody works together. In this quote, she highlighted the importance of central office and the curriculum office in particular making the decisions regarding curriculum. This belief that district leadership drives change was also evident in the Hillside DN meetings. At these meetings, Ms. King provided the agenda and primarily delivered content to members with some opportunity for discussion. She would often have Blueprint concepts for members to discuss in small groups and then share their thoughts with the group. For example, at the December 6th meeting, Ms. King talked through survey results regarding if teachers felt supported by their school and district leadership. In this meeting, she talked through the results from different survey questions and facilitated some discussion around how each school’s BNs felt about these survey results. Overall, most of this meeting was led by Ms. King with minimal input from DN members. As the district worked through the “phases of installation” of the Blueprint, Ms. King worked to ensure that the district leaders were adhering “with fidelity” to the directions from MI Excel. At the beginning of her work with the Blueprint, Ms. King made decisions and established 93 routines to ensure that Hillside followed the directions from MI Excel with fidelity. First, upon signing onto the program, she and Ms. Nelson, the principal at Willow Elementary, also enrolled into the facilitator training for the Blueprint. Traditionally this training was for members of Intermediate School Districts (ISDs) who would push into school districts to support their work with the Blueprint. This training ensured that they would have advanced knowledge of the Blueprint program and follow it. Additionally, Ms. King established a close relationship with the creator of the Blueprint, Dr. Grant Chandler. She repeatedly invited him into the district to present to district staff about the Blueprint. She also had a standing weekly phone call with Dr. Chandler to discuss the district’s progress. Weekly, Ms. King, Ms. Nelson, and their ISD representative called Dr. Chandler after DN meetings. Ms. King summarized these phone calls as a way to help him guide them through the program: Every Wednesday morning after we leave the DN and we talk to him about what we’ve done and what the next steps need to be. And then there’s steps in the DN about where you need to go. And we talk to him about where we need to go afterwards, so he really guides us through that. Collectively through the additional training and communication routines, Ms. King ensured that at Hillside they were following with fidelity the Blueprint program as designed by Dr. Chandler. She described these routines as making sure that she “was on the right track” with the district’s Blueprint work. When considering her schools, Ms. King focused on ensuring fidelity from her school leaders and teachers to the systems and programs designed by the DN. She saw the BNs as ways to implement DN programs at schools. She explained, “well the BNs carry out the work of the DNs in the building. So, the DN set the role and the goal and the BN did the work at the building 94 level.” Beyond this expectation, she ensured that the BNs were completing this work through collecting copies of the agendas and approving the goals of each BN in her district. The principal at Hickory described the struggle to get her goal approved by the superintendent: We were told to devise goals. But we were told and everybody thought the same thing, again sometimes the communication is not really very clear. We had gotten data. Remember the student surveys we talked about? So we were given that information then we were told to do the goal, so everybody thought that you need to and you had to go through the problem solving protocol, so everybody was using that data….We all handed in our goals…But anyways we had to do the goals over, every single school. …like it was obvious that they were not what they were looking for, so now -we’re still, we should have the goal done by the next meeting. From the district’s viewpoint, approving BN goals ensured faithful program implementation to the goals set at the DN. However, as seen in Table 4.1, while 84% of the Hillside superintendent’s comments reflected an experimental science perspective, she did not exclusively reveal this perspective. In three instances, she did share viewpoints that aligned with the improvement science paradigm. All three of these quotes pertained to a willingness to modify the systems and programs created by the DN. First, Ms. King described needing to alter how principals completed their high- quality walkthroughs. At the beginning of the year, principal were not supposed to give feedback to teachers based on the HQI walkthroughs; this was simply a routine to collect data on teacher instruction. However in April, they began to give feedback to teachers. This modification was made in response to teachers asking for information on what principals were seeing in the classroom. Second, she indicated that the designated curriculums that teachers were required to 95 teach (as described further below) would be revisited in the early summer by the curriculum committee and changes would be made based on feedback. It is important to note that these revisions were only allowed to occur after the school year ended and all of the units had been taught. Overall, Ms. King described more experimental science assumptions regarding leading district change. She outlined the change process as where district leaders created plans, while school leaders and teachers followed the plans. She prioritized fidelity both to MI Excel in how she envisioned her work in implementing the Blueprint and fidelity in implementation from school leaders and teachers as they worked to implement the systems designed by district leaders. As the school year progressed, Ms. King did express some improvement science beliefs in allowing for some modifications from school leaders and teachers to some previous decisions. 96 Hillside Theory of Action for the Instructional Infrastructure Figure 4.3: Hillside School District Theory of Action for Influencing Reading Instruction At Hillside, the theory of action to influence reading instruction at scale reflected Ms. King’s assumptions about change. During the study year, Hillside school district was in its 97 second year of implementation of the three total years that Blueprint implementation was expected to take. As shown in Figure 4.3, overall, Hillside followed the recommendations of Blueprint leaders in how to influence Reading instruction at scale across the district. First, district leaders designed the vision of high-quality reading instruction. Then they began the process of designing the curriculum. Ms. King described this process as district leaders surveying teachers for their preference with the result of, “Many of them said they wanted to go back to a traditional book, so we began.” They ultimately chose the Journeys textbook series. District leaders then organized a committee to write the curriculum drawing off the Journeys text and the vision of high-quality reading instruction. Ms. King described the process of writing and sharing the curriculum as: A group, anyone who wanted to participate on the group was able and then over the summer, we emailed out and asked for people that wanted to help write. Two people from each grade level were selected to write all summer long. And then in the fall we had workshops and so there was an institute day. On December 7, 2016, I observed the science curriculum committee working to write Overall, the Journeys textbook and district reading curriculum written by grade level teachers served as the mandated curriculum. Additionally, all teachers attended professional development known as “curriculum roll-out meetings” to learn about each upcoming unit. their unit 3 for district teachers. This committee meeting occurred in a spare room at the central office building. At the meeting were teachers who served as members of the committee and a Blueprint leader who approved units as the teachers completed them and provided feedback as they worked. Ms. King also regularly stopped in to check the teachers’ work and provide 98 At the differentiated instructional improvement network level, Hillside created a data feedback. The presence of a Blueprint leader and Ms. King at these meetings helped monitor the fidelity of the curriculum to the vision and other district expectations. management system where teachers were required to submit the number of students who earned at least 90% on their unit test. This data was analyzed at PMT meetings. Additionally, during the 2016-2017 school year, Ms. King hired two literacy coaches to support teachers’ instruction and implementation of the curriculum. Because the study year was in the second year of implementation, opportunities for teachers to collaborate and individualized professional learning systems were not yet implemented; however, by the 2017-2018 school year teachers were to have common planning time built into the master schedule. provided by the Blueprint with fidelity. They did this by having Blueprint leaders present at curriculum development meetings and through regular communication with Blueprint leaders for approval of their decisions. As described above, Ms. King held experimental science assumptions regarding change and prioritized the importance of implementing the Blueprint with fidelity across her district. The reading instructional infrastructure system provided one example of how she faithfully implemented the Blueprint framework into Hillside School District. Overall, Hillside leaders followed the theory of action for influencing instruction Mr. Dixon – “Classrooms should be a petri dish” Mr. Dixon, the superintendent at Lakeside had a different path to the position of superintendent. The majority of his career had been at Lakeside. First as a 5th through 12th grade band director, then as an elementary school principal, then as a high school principal, then in human resources, and finally as superintendent. 99 Like Ms. King, the superintendent at Hillside, Mr. Dixon decided to undertake the Blueprint program during his first training as well. He described it as an easy decision, “I barely knew Grant, but if Grant was going to come and try to tell me to change a core value that I believed in, I was going to tell him no. So I wasn’t afraid to say yes.” In other words, when saying yes to Grant, he knew that he was going to modify the Blueprint to meet his needs and the local specifics of Lakeside. He later communicated how he appreciated that the Blueprint was not simply a canned or “cookie cutter” program that he was being asked to implement: And that’s to me, that’s actually what I loved about the plan is that it’s, you know, it’s Apple versus Android. Android people develop their own apps. It’s a free platform. I’d prefer that method than someone telling me this is the app and I have no say in what the app’s going to do. You know here I look at all of these like on the Blueprint just like an app. You know it’s something we’re putting out there, people use it. Let’s make sure it’s something that is useful if they’re going to use it. Let’s make sure it’s useful and does what they need it to do. This willingness to develop the Blueprint to meet the needs of Lakeside was clear in DN meetings, where not only did they complete planning tools, they also altered MI Excel created “Evidence of Practice” frameworks for the systems they were implementing, so that the language in this tool matched the language used by staff at Lakeside. As a leader, Mr. Dixon described himself as the decision-maker, who acted after hearing the viewpoints of the members of his team: So, so to me, I’m privileged to be in a position where I get to be the loud voice, but you know with the team that I have, which are an amazing group of people, I can challenge them just as hard as they challenge me. And so, it’s definitely not a tyranny here. It’s 100 definitely not a you’re going to do what I say. At the end of the day, they live with the decision that I make, but I hope that they would say this, that the decisions that we get to, are made because of the group discussions we have. This ability to challenge and be challenged was evident in the district’s DN meetings. Many meetings involved discussions in which the members asked questions and challenged each other’s ideas. For example, throughout the year, the Lakeside DN struggled with the Student Support Network (SSN). At each meeting, the chair of the SSN committee provided an update on the committee’s work. By the March 28th meeting, it became clear that she had a different vision than the rest of the DN. She envisioned a sort of monthly hour-long lesson for junior high and high school students. The DN members worried about rolling out this curriculum next year with new ELA and science curriculum. The special education coordinator worried that this vision of the SSN might be too broad, when an easier focus for the upcoming year could be in (1) checking in and retraining on their pre-existing PBIS system and (2) working to implement MTSS. The debate continued for a while and ended with the superintendent asking the SSN committee to continue working on their curriculum, but with an understanding that it may not be implemented in the next school year. These debates and discussion at DN meetings had a very different feel than the DN meetings at Hillside in which the superintendent directed the conversation and shared information from MI Excel. The vast majority of Mr. Dixon’s interview quotes fell under the assumptions of the improvement science paradigm, because of (1) his belief in his people, (2) his willingness to revise and modify regularly DN established systems and procedures, and (3) his interest in variation. 101 First, Mr. Dixon described the strength of Lakeside as being in its people. “You know, I am asked so what is one of the best assets of this school district? And I say all of the time it’s our people.” He continued to say that the people in his district were fighting for the common good and while they may disagree, it was never about the work that they were doing, but that they challenged each other to create the best possible system. Mr. Dixon explained, “Nobody’s fighting the implementation of the system. That’s a huge difference, huge difference No one’s saying ‘No I don’t want to help kids.’ You have, ‘No, we’re not…that’s not helping kids enough.’ That’s a whole different direction and a different argument that’s happening at our table across our district.” This belief translated into how he envisioned the implementation process under the Blueprint: I think that there’s a lot of good people out there that just really need advanced, intensive training. If you give them that training and you get the right person, that’s the combination that wins. Right there. That’s when you’ll be ok. So for us, again going back to what I said, I think we have amazing people. So now it’s us doing our job and supporting the and making sure they grow. As further discussed below, Lakeside’s plan to improve instruction at scale included keeping the systems that some teachers were successful with and training the rest of the staff so that they had better understandings of the literacy programs and how to use them. Second, Mr. Dixon and his DN members often discussed their plans to refine the systems they were creating. As they rolled-out new programs into their district they acknowledged that they were just at the starting point and that things would change. The superintendent conceptualized their Blueprint work like creating a fine whiskey: 102 I’ve always compared it to whiskey. You know, as you run a whiskey through the refinery every time it comes out more pure and it comes out stronger and so you know the reason that you have a top shelf whiskey or a top shelf vodka is it’s literally gone through the refinement process so many times, it’s like the most purest and most high quality form of that alcohol you can get. And I think we’re trying to do the same thing with us. We’re trying to create a top shelf system and how you do that is I don’t think you just do it once. You refine it. You continue to run it through the same protocol every single time. …But now what we have to do is refine the ones that we have put in plan and make that next level better, because you know, you hear that there are concerns with some of things we’ve implemented that-they’re not jugular concerns, but they’re things that we know are that are standing in the way of us doing the best we can. So we need to refine them and put them through 2.0. And then eventually 3.0 and 4.0…That’s just where you’re trying to work towards is that you have something that is just a high-quality instrument and we think that it’s pretty good quality out of the gate, but we know that we can make it better. In this quote, Mr. Dixon conveyed his willingness to refine the systems that the DN created. This ongoing refinement was best represented by the process for principal HQI walkthroughs. Throughout the 2016-2017 school year, administrators at Lakeside worked through three different data collection and feedback forms for the HQI walkthroughs. The superintendent admitted in August that in response to inadequate data collection from the systems that they had created that they had to “start over and do it again.” This willingness to refine was also present in the PSP system, which was revamped often during the school year in order to help teachers have more productive and focused conversations around their data. 103 Third, the Lakeside superintendent appreciated the feedback that variation in implementation allowed the DN to receive in their PMT meetings and to hear about the practices in different schools: We’re getting ready to do our third (Performance Management Meeting) here in December and so that is an opportunity for us to be able to hear from them [the principals] what’s going on in their buildings and be able to get some, you know, be able to see differences in the buildings, be able to see similarities, be able to create bridges between buildings of “Hey. You’re doing this well, this building’s not. You guys should talk. Figure that out.” And also for us to be able to say “How are we doing as a district?” That aggregate level stuff. So that’s very, very helpful. In communications from the principals, it was clear that some of the elementary principals learned from the practices of other principals in these meetings, as each principal worked to implement these systems in his or her school. For example, the principal at Cottonwood used a form in her PSP in May that was developed by staff at Lilac Elementary School. She learned about this form at an earlier PMT meeting. As will be further described below, principals at Lakeside were given the freedom to create these forms, which would not have been the case at Hillside. There were four times that Mr. Dixon did convey perspectives aligned with the experimental science paradigm. All four of these came in reference to wanting fidelity to the literacy instruction systems that the district had put into place. In describing the HQI walkthroughs, he admitted that there were understandable times in the day, when a teacher would not be engaging in high-quality instruction, such as in the five minutes getting ready for gym. 104 However, he did believe that these walkthroughs could help monitor for instruction aligned with their vision: On the other end though, if I go into a room 4, 5, 6 times at various times and I can honestly say in none of those times and none of those visits, I’ve seen any kind of high quality teacher or student action, that’s a concern. …you know and what’s going on and have that discussion and get them to start focusing on maybe I’m not using my time as efficiently as I could. While in this quote and two others, he does convey a desire for teachers to teach with fidelity to the HQI visions and the literacy systems (CAFE and Daily 5), he expects this fidelity after taking time to listen to teachers and build consensus around the goal. Overall, Mr. Dixon primarily expressed beliefs consistent with the improvement science paradigm of change. He was supportive of making modifications to district-designed systems throughout the school year based on user feedback and data. He also shared a belief in the value of the people working in Lakeside School District. He believed in their expertise and ability to effectively do their jobs. Lastly, Mr. Dixon allowed for variation and welcomed the new information that his petri dish classrooms could provide to the district regarding effective instruction. 105 Lakeside Theory of Action for the Instructional Infrastructure Figure 4.4: Lakeside School District Theory of Action for Influencing Reading Instruction As shown in Figure 4.4, the Lakeside theory of action for influencing reading instruction at scale throughout the district both followed the theory of action created by the Blueprint and reflected some of Mr. Dixon’s improvement science orientation. First, leaders at Lakeside followed the direction in the overall structure of their theory of action. They created a vision of 106 high-quality reading instruction and made it available on the district website as well as at professional development. Second, district leaders chose curriculum for the district. Like at Hillside, district leaders consulted with teachers. However, unlike at Hillside, they ultimately decided to mandate throughout the district the instructional practices that the teachers at Cottonwood Elementary School were using successfully. They chose the Making Meaning textbook because teachers at Rose Elementary School already had experience with the book. Unlike the directions from Blueprint leaders, Mr. Dixon and his district leaders did not create or mandate a pacing guide or curriculum to go along with these resources. In order to ease staff stress in this second year of implementation, district leaders really stressed training teachers on how to use these systems without any additional expectations. Mr. Dixon summarized the focus as: So right now, we’re only doing Reading. We took some systems that we already had in place like guided reading, like CAFE, using DRA assessments in that way, and those are systems that we’ve had in place for probably about 10 years the problem is that we did none of them with fidelity. So we retrained everybody and now even people that will say … that “I have done CAFÉ for 10 years are now saying this is actually my first year doing CAFÉ.”…So, we’re trying to make sure that we get everybody to implement it the way that those tools and those systems and those resources were developed. As shown in this quote, Mr. Dixon and his staff decided to use the second year of implementation to really invest in his teachers through additional training and a focus on embedded training from principals. This focus on people reflected the improvement science assumption that improving instruction at scale depends on the knowledge of the people (Lewis, 2015). 107 Like at Hillside, not every component of the Differentiated Instructional Improvement Network system was implemented in the second year of Blueprint implementation. In terms of data collection responsibilities, teachers were required monthly to submit a reading level based on their DRA assessments with each student. These DRA scores were a component of teaching under the CAFE literacy system. Teachers were also provided time to collaborate around data at monthly PSP meetings where they would meet as grade level teams to analyze student achievement data from the past month, which could include reading level or NWEA scores. During the 2016-2017 school year, Lakeside had not yet introduced full-time instructional coaches to assist teachers. Some teachers served as instructional coaches to their colleagues, but their role was limited since they were also responsible to their own classroom. This changed for the 2017-2018 school year, as the district transitioned away from instructional aides to providing a full-time literacy coach to each elementary school. Overall, Lakeside leaders followed the theory of action for influencing instruction provided by the Blueprint with some modifications reflecting the improvement science orientation of Mr. Dixon. They did this by modifying the curriculum development and approval process into a two-year process. In the study year, leaders prioritized investing in teacher knowledge and experience with the CAFE and Daily 5 systems. In the 2017-2018 school year, they planned to introduce a curriculum and pacing guide. As described above, Mr. Dixon held improvement science assumptions regarding change and highlighted the importance of building knowledge in the people when mandating a literacy program. The reading instructional infrastructure system provides one example of how he modified the theory of action provided by Blueprint leaders to fit his local context and beliefs. 108 Implementation Orientations at the School Level The superintendents’ perspectives on the implementation process had great influence on the actions and beliefs of their building principal team as they worked to implement the Blueprint systems designed at the district level into their buildings. In this study, I interviewed two principals from each district to understand how the Blueprint implementation process was happening in their schools. Interestingly, most of the principals expressed more improvement science assumptions, with the exception of the principal at Hickory in Hillside. Each of these principals worked in their own unique manner to implement district-designed systems into their schools. Their strategies were informed by the beliefs and pressures from district central office and their own beliefs on how to enact change. Interestingly at Hillside School District, one principal shared more experimental science viewpoints while the other communicated more statements aligned with improvement science. At Lakeside, both principals shared more improvement science assumptions, but focused on different parts of the framework. I begin my analysis by describing Ms. Davis, who was the only principal to communicate more experimental science assumptions. I then analyze the three principals communicating more improvement science assumptions. Ms. Davis – “Walk that line” Ms. Davis was in her first year as a principal at Hickory Elementary School after serving as a teacher in the district for over 20 years as a 1st and 2nd grade teacher. During the 2016-2017 school year, she communicated a desire to build relationships with her staff. At the same time, she wanted to comply with district mandates in terms of the Blueprint requirements, because she honestly believed in the program and that the structures and routines would improve teacher and student performance at her school. 109 In trying to follow district directions, she communicated that, “We’re really told a lot what to do and then sometimes we’re not supposed to do things, so like it’s really hard to walk that line to know where, when I’m supposed to and when I’m not supposed to.” Throughout the course of the school year, she repeatedly tried to implement the systems as designed by the district only to receive more information from district leaders asking her to revisit the work that she and her BN had done. She discussed how her BN meeting went when the ISD facilitator visited: So then they evaluated our BN, [ISD coordinator] came and evaluated us or analyzed. She’s like, “Well how many times have you done the protocol?” I’m like “We, we’ve done it once. We started in January. We handed in goals. You said the goals sucked. We did it again. We used the same data to come up with a different action plan. You didn’t like that.” So we came up with our last action plan, which was vocabulary and that didn’t fly either. During the fall interview, she seemed to be frustrated by the district restrictions she faced in implementing these systems into her school. She shared that in terms that “I know what I could make it, want it to be, but you know you’re supposed to leverage the district systems.” Throughout the school year, Ms. Davis was stuck in the middle, as she tried to implement these systems exactly as district leaders want them, while making them authentic for her practice and her teachers’ work at Hickory Elementary School. her role as, “Just enforcing what was already written, really making sure the curriculum is being followed.” This role prioritized the expectation that teachers implement with fidelity the program created by the district. Additionally, Ms. Davis accepted as her role as a principal at Hillside to In terms of acting as an instructional leader for the reading curriculum, Ms. Davis saw 110 monitor fidelity of implementation. These orientations toward changing instruction followed the logic of experimental science (Lewis, 2015). As she made plans for the 2017-2018 school year, she figured that the solution to these misunderstandings with the district was to increase her communication with district leaders, so that she could better comply with district directives: I never, I didn’t feel comfortable with BN not really understanding and again knowing, because we really had to please central office, it was top down, even though it’s not supposed to be. So trying to, to help this, to grow this team of people into a real team that is really trying to be a force of change, instead of complaining and compliance. You know like what are we going to do? We can do this if we’re a team, but we have to, we really have to be on the same page. She communicated that in the future she wanted a sort of “Batman phone” to central office, so that central office could approve her plans before she did them. This way she hoped that she would not have to go back to her staff to re-do something that they had done incorrectly, such as the BN goals or the PSP action steps. Basically, her plan was to gather more information so that she could more faithfully implement district programs into Hickory Elementary School. The few times that Ms. Davis did express a more improvement science orientation to change was in terms of allowing for modifications. Ms. Davis felt that modifications to the reading curriculum and data collection process should happen over the summer. In her June interview, she described the changes that the reading curriculum writers made to the data collection process at the beginning of the summer as, “curriculum writers know, they're going to highlight. Like the ELA people said this is what we're going to do. We're going to do this on week three, this on week five, then we’ll have this.” A difference between the Lakeside leaders’ 111 and Ms. Davis’s view of modifications seemed to be in when they felt that considering and implementing modifications was appropriate. Ms. Davis and Ms. King both communicated a belief that the summer was the appropriate time to consider modifications to the Reading curriculum. At the December PMT meeting that I observed, the administrators had already noticed this problem with their data collection, however this solution was not implemented until the reading units were revised over the summer. Overall, Ms. Davis prioritized compliance and fidelity in both her actions as a principal implementing district-designed systems and programs into her school and in terms of her teachers’ reading instruction aligned with the district-designed curriculum. This focus on implementing with fidelity was aligned with an experimental science orientation towards change (Lewis, 2015). Ms. Nelson – “I understand it on a deeper level” Ms. Nelson was in her second year as principal at Willow Elementary School. Interestingly, while she led a school in the Hillside School District, in her interviews she communicated more improvement science assumptions about implementing the Blueprint than experimental science assumptions. This may be due to the fact that unlike Ms. Davis, she had been going through the Blueprint facilitator training with Ms. King and their ISD facilitator. She shared that this additional training had given her a deeper understanding of the Blueprint as well as made her a leader on the district level. This additional knowledge might have been what made her feel like she had more flexibility in how she implemented the district designed Blueprint systems into her school. With her deeper knowledge of how the systems were supposed to function, she did not need to double- check with district leaders regarding her BN and PSP processes. For example, when observing 112 her BN, it was clear that their goal was around improving student-teacher relationships. Ms. Nelson shared that one of the goals of her BN was “to improve the way that students felt that teachers and staff cared about them in the school.” This goal stands in contrast to Ms. Davis’s experience in which she was told to change her BN goal away from a teacher-student relationship goal. Overall, this additional knowledge and flexibility might have contributed to Ms. Nelson implementing Blueprint systems with more improvement science assumptions in her school. First, she and her BN regularly completed PDSA cycles in their meetings by using the PSP protocols. For example, in their March 28th meeting, taking their relationship goal, they studied survey data from question regarding if students “felt like their teachers knows about them or cares about them”. In the results, they noticed that their high-achieving girls overall felt less like their teachers cared for them than their peers. As Ms. Nelson described, the BN then took action: When we collected our second set of data that’s when we saw there was a drop, even though we went up and we met our goal, there was a drop in the area of high-achieving girls. … They’re not getting called out for anything because we’re looking at those behavior problems and trying to get them on track. And you know so we kind of identify that as being a concern and since it was kind of girls overall, but mostly high-achieving girls, we decided to focus on the girls for the “Paw-some Notes.” We thought this would be a way different professionals in the building could interact, like I know who you are, I care about you. The idea of “Paw-some Notes” came from the BN. Over the next week, BN members put a list of girls and blank “Paw-some Note” sheets in the teachers’ lounge. Then while teachers were eating lunch, if they chose, they were to write a note to a student that they knew, but who was not 113 currently in their class and tape it onto that student’s locker and cross the student’s name off the list. While they ran out of time in the school year to collect another round of data on the survey, Ms. Nelson commented that these notes resulted in positive morale among students and parents in her school as some parents even posted on Facebook about what “their child’s teacher wrote.” Beyond engaging in PDSA cycles, Ms. Nelson showed improvement science assumptions in how she worked with her teachers regarding the changes to their ELA instruction. She admitted that there will be variation in how teachers implemented the curriculum and consequently in how she responded to a teacher’s instructional practices. She shared that she saw her role less as a faithful implementer of district mandates, but more as responding to the needs of her teachers as they worked to teach to the new guidelines. Ms. Nelson explained: It’s a delicate dance. I know who I can push a little more, and who I can’t quite push that much. I’m learning that in the second year, but that has been a struggle, obviously. Some teachers, it’s a real shift from how they used to teach, and so that, in and of itself, is just a struggle. My goals for each teacher are probably a little different. What I see as progress in one classroom, would not be progress in another. I’m just trying to keep the—be mindful. This acceptance of variation in implementation based on the people in her school reflected the improvement science assumption that people matter when improving instruction at scale (Lewis, 2015). In summary, while Ms. Nelson was a principal in the Hillside School District, she presented more improvement science assumptions regarding change than experimental science assumptions. This improvement science orientation was evident in her regular incorporation of PDSA cycles in her BN meetings and in her acknowledgement of variation in practice based on 114 the backgrounds of the people in her school. Her advanced training in the Blueprint could have contributed to her improvement science orientation by providing her with an understanding of the theoretical foundation of district mandates. Ms. Wilson – “Doing it with purpose” In the 2016-2017 school year, Ms. Wilson was a first-year principal at Cottonwood Elementary School. Prior to taking this position, she was a high school English teacher at Lakeside High School. She often connected her work as a principal to her experiences as a teacher and what she wanted from her administrator when she was a teacher. Overall, she made the most improvement science statements among all of the participating administrators. It was clear that she held strong improvement science assumptions, especially around the areas of building consensus around the systems and programs, mapping the rationale behind the systems and programs, and refining programs. First, Ms. Wilson focused her Blueprint work around building consensus and being transparent with her staff about the reforms. She remembered that as a high school teacher, when the Blueprint was announced, there were a lot of conspiracy theories: “Everybody thinks that everything’s a conspiracy…What’s going on? Who’s pushing this? Is it a district thing? Is it a state thing? What’s their intent? Are they getting rid of us?” Drawing off these memories, she described her goal in implementing the Blueprint of getting everyone on board. She defined the mission of a leader as “to get people to buy-in to what you’re doing.” She felt that the easiest way to get people on board was by giving them some ownership of the Blueprint at Cottonwood Elementary School: I felt like it was so important for them to not only understand it but to make it theirs. You know and to make it something that was our school’s or that was cutting edge, you know. 115 Like I often said that like “we're doing this and I'm sharing this out with the other leaders” and people are saying well when you figure something out that works, we want to use it too. I feel like they felt like pioneers. They did that with CAFÉ, and so naturally I have some really strong leaders on my staff which totally bought into that. But you know, it doesn't make any sense if one you don't understand it. Right? If you don’t know what you're doing, then why do it, because it doesn't make sense. And two if it doesn't have a purpose then why are we doing it?” Ms. Wilson worked to accomplish this goal of getting her staff to buy-in by using staff meetings to share the things that the BN was working on and letting BN members take leadership roles regarding different systems of the Blueprint. For example, Ms. Parrish led Cottonwood’s communication driver. She collected data regarding the school staff’s usage of the protocol, shared it with the staff at meetings, and set new goals for the staff. Ms. Wilson believed that through transparency and an increased role for teachers, her staff would buy-into the program. Ms. Wilson described that her primary goal when it came to the Blueprint was “to help people understand how it all fits together”. It was really important to her that they understood that each system was not just one more thing, but that it was all connected: So I think that’s a huge positive that people were really receptive to trying new things and I also feel like we really immersed ourselves, they did, in understanding the Blueprint. And like why do we collect data and how do we use that data and how does that drive our day-to-day instruction, and you know, just tie everything and really streamlining it together. Like I think before they were doing things because they just knew that they had to and maybe the previous principal wasn’t really comfortable with what those were and so they just did it because they were told to as opposed to doing it with purpose. 116 As described above, she took a lot of time in their monthly staff meetings to go over the Blueprint. Additionally, she created a voluntary “High Quality Reading Study” staff learning opportunity in which teachers could come and learn more about the HQI walkthrough form that she was collecting data on during her walkthroughs. At this meeting, the teachers had the opportunity to ask questions as well as act as a principal by watching a videotaped lesson together and assessing it using the HQI walkthrough form. She described the conversation in this meeting being a huge positive for the staff’s understanding and buy-in. Lastly, Ms. Wilson demonstrated her improvement science orientation in her willingness to modify and refine tools provided by the district to meet the needs of her staff at Cottonwood: So if we're just doing PSPs because we were told to do it, that doesn't make any sense. And unfortunately because of the way that things were given to us a lot of ways it was easy to get very caught up in the process. So like PSPs were something, we struggled with all year because we would do them and we would spend an hour on the causal statements. You know but like what is the main point of a PSP? It’s the action plan. You know so it’s wasted time and so for us it was like OK how can we refine our process to get what we want to out of it? And that's why I feel like we really did the work this year, because if we take the work that we've done with our PSP and with our forms for the SST and you know we take all of that and we use it for next year I really feel like we can accomplish and do all the things we want to do. Her BN also took a leadership role in the district in refining things given to them by the district. For example, over the school year, I observed three PSP meetings, and in each meeting, teachers were using a slightly different tool to analyze the data. These different tools were designed by the Cottonwood BN in response to struggles in the previous PSP meeting in regard to the tool 117 taking too much time or not focusing on action steps. Overall, BN members and the principal agreed that the tool that they used in May was a successful one that they would continue to use in the following year. Even with her strongly held improvement science viewpoints, Ms. Wilson did take some experimental science actions during the school year. Most of these revolved around making sure that the ELA program was taught with fidelity. Early in the year, she and her BN made a plan to incentivize teachers to teach in ways aligned with the CAFE and Daily 5 literacy programs. Incentivizing implementation is a component of the experimental science program. Their plan was to reward teachers through a program where when she observed two high-quality teacher or student actions on the HQI walkthrough rubric, teachers would get a raffle ticket for a monthly raffle. Ms. Wilson described this plan as a “huge miss for us.” Because what ended up happening was that her teachers went to the union and it became a district issue. However she shared that she really reflected from this incident and from this step back, it seems like she became more intentional in her improvement science practices. She focused more on the purpose behind her actions and making them transparent to the teachers, “I missed the mark with the ticket because I'm like I would be happy with a ticket as a teacher, but it's like well yeah but it does that really grow high quality instruction. No because they're not getting any feedback.” Overall, Ms. Wilson conveyed strongly held improvement science assumptions about change. She more than any other administrator prioritized the importance of building consensus around changes and creating learning opportunities for her staff to understand how the systems mapped together. Like Mr. Dixon, she led efforts to modify elements that were not working for her staff at Cottonwood Elementary School. 118 Ms. Reed – “Not just our school but our district” Overall, Ms. Reed was Lakeside School District’s turnaround principal. She began her career at Lakeside as the middle school principal, after the middle school had been placed under priority status. After the middle school exited the list, she transferred to the elementary level for the 2015-2016 school year, when Rose Elementary was placed on the priority list. She admitted that she was still learning how to be an elementary level principal and that she had been noticing differences in working with elementary teachers rather than secondary teachers. In addition to her role as the school principal, she served as the only elementary level principal on the DN in the 2016-2017 school year. She brought her experience as a turnaround leader to the Blueprint program. From her experience, she shared an improvement science belief in the need for staff buy-in: One thing I learned myself as a principal as a leader of turnaround is the more informed people are, the more on board they get. And what people don’t like, educated people, professional people, who are vested in their positions, in their job, they have a passion for what they do. So I think the ineffective teachers, are people who are not so passionate and are the opposite, “Tell me what to do and I’ll do it.” Right? But people in order to get good people invested in doing, because you can’t in turnaround, it can’t just be something given to you. You have [to] actually internalize it and feel that passion for it. Ms. Reed worked in a couple of ways to get her staff invested in the Blueprint. First, she had a motto in her BN meetings of “have fun and get shit done.” This motto meant that her meetings tended to be laid back and have very open conversations. There was frequently laughter among the group. Second, because of her role as a DN member, she regularly began every monthly staff 119 meeting with updates from the district level. She asked for feedback from her staff to take back to the district level. Ms. Reed while also demonstrating improvement science assumptions in her stance towards the Blueprint, qualitatively was different from Ms. Wilson. While Ms. Wilson’s improvement science actions seemed based in a desire for her staff to understand the program and buy-into it., Ms. Reed focused most on a desire to make revisions. This stance seemed primarily rooted in her role as a leader at the district level. Throughout the school year, she used the information gathered in her experiences at her school to try to make the systems better for the entire district. She took leadership around improving a couple of district-level systems. First, she described using three different data forms for HQI walkthroughs throughout the school year, first through a Google form, then a tally system, and finally a feedback form. In her role as a DN member, she was in regular communication with the district’s director of instruction and the technology expert to work to revise these forms for all district principals. Second, she brought issues from her building to the DN. Most notably, she brought the behavior issues she was having in her building to the DN and this resulted in a change to the curriculum expectations for the 2017-2018 school year. She described noticing this problem and feeling a need to bring it to district leaders’ attention: What we found was we were pushing the instruction and academics so hard in September and October, that we actually, we felt like we didn’t go over procedures, and we didn’t go over rules and expectations and we jumped right into math. Our math scores actually looked good, but the behavior piece needs to… because I heard across the whole district it was going batshit crazy everywhere but so then I knew it wasn’t just me....So that’s why I brought it to them. What are we doing? 120 From her raising this concern at the April 11 DN meeting, district leadership altered their instructional expectations for the beginning of the 2017-2018 school year at the elementary level. They also introduced different expectations for the kindergarten level, so that these teachers could have a prolonged period of time to establish rules and routines with their students. Lastly, she was a leader in redefining how the Lakeside DN worked by introducing roles and responsibilities for each DN member and district administrator. She introduced this idea at a MI Excel training to the DN, because she wanted more administrators to feel ownership of the Blueprint. She described the system as: We now have roles so like all, every administrator in the district, not just DN, but every administrator has a kind of specialty like mine’s BN. So if it’s a resources for any principal, hey if they’re struggling with BN, I might go to theirs, help them out, look at their agendas, they can look at my agenda. They were planning on implementing this in the 2017-2018 school year with the hope that this system would allow more administrators to have a deeper understanding of at least one Blueprint system. As a principal on the DN, Ms. Reed used the improvement science mentality built at Lakeside to push for modifications to systems not only in her school level, but also at the district level. Like the other administrators, Ms. Reed also had comments that aligned with the experimental science paradigm. Specifically, around wanting to monitor fidelity of implementation. However, her concern was not about teachers in her building implementing programs with fidelity, she was concerned with principals at other buildings being monitored and getting feedback on their implementation of the Blueprint. 121 Overall, Ms. Reed like the other administrators at Lakeside exhibited an improvement science orientation to change. She communicated a belief in the value of building consensus and buy-in around the changes being made in the school. Most of her improvement science focus was around modifying and refining systems that had been implemented based on her experiences and feedback from her staff. Different from Ms. Wilson, Ms. Reed’s focus of these modifications tended to be based at the district level rather than what was occurring within her school. School Principal Summary Each principal demonstrated through their actions and interviews that they had different beliefs regarding implementing the Blueprint systems and district programs into their schools. Their beliefs and actions seemed to be influenced by pressures at the district level. However, each of them worked in some way with their knowledge of their local environment and relationships with their staff to modify these systems and programs to work in their buildings. Ms. Davis was unique in that she seemed to feel the most beholden to following district guidelines faithfully. This was probably a consequence of being new to her role as a principal as well as the more experimental science viewpoints of her superintendent. Ms. Nelson did not seem to feel as beholden to district guidelines and instead used her deeper knowledge of the Blueprint to incorporate her improvement science assumptions into the change process at Willow Elementary School. Ms. Nelson regularly utilized PDSA cycles in her PSP meetings and allowed for variation in instructional practices, understanding that teachers will be at different levels of implementation of the new systems. Both principals at Lakeside, Ms. Wilson and Ms. Reed, primarily exhibited improvement science assumptions around change. Ms. Wilson focused primarily at building consensus and understanding around the new systems in her staff. Additionally, she and her BN took a leadership role in modifying district systems to better fit the 122 needs of Cottonwood teachers. Ms. Reed focused more at a district level and incorporating information from Rose Elementary School into the conversation at DN meetings to inform modifications to changes at the district-level. Chapter Summary This chapter explored concepts related to my first two research questions: (1) What is the intended theory of change of the Blueprint for Systemic Reconfiguration for influencing instruction at scale? and (2) How does each administrator’s orientation toward change influence the implementation of Blueprint for Systemic Reconfiguration in their district or school? Overall, the foundation of the Blueprint for Systemic Reconfiguration was built on improvement science assumptions of change. The Blueprint prioritized systems reform, the use of data, and local decision-making to design programs and routines customized to the local environment. While the Blueprint was built on improvement science assumptions, district and school leaders brought their own backgrounds and orientations toward change to the implementation process. For Ms. King and Ms. Davis, their experimental science assumptions towards change resulted in a focus on faithfully implementing the Blueprint program into the district and district programs into Hickory Elementary School. Administrators with a more improvement science orientation towards change, such as the leaders at Lakeside and Ms. Nelson, developed and welcomed modifications to the Blueprint in order to fit the needs of the people and circumstances in their district and schools. In the next chapter, I will explore further how these orientations towards change influenced specific decisions regarding the process to influence teachers’ reading instruction practices. 123 CHAPTER 5: IMPLEMENTATION INTO SCHOOLS AND CLASSROOMS This chapter seeks to answer my third and fourth research questions: (3) How is the theory of change enacted in each district? How does this enactment reflect administrators’ orientations toward change? and (4) To what degree and in what ways does the enactment of the theory of change influence teacher instructional practices and planning in each school? Leadership styles and assumptions were manifested in two ways in the enactment of the theory of change and implementation of district instructional policy through the Blueprint: (1) the literacy program that was mandated and (2) how leaders enacted their literacy instructional policies. In this chapter, I explore how leadership assumptions about change were reflected in the programs they chose to mandate and the implementation systems that they installed to monitor and support teachers’ implementation of the programs. Additionally, these administrative choices seemed to influence teachers’ perceptions of the Blueprint reforms and their willingness to work to change their instructional practice. Teachers at each district enacted the expectations differently based on their perceptions of the reform, their previous experiences, and their knowledge around reading pedagogy. Mandated Literacy Instructional Policies Both districts, for the most part, followed the theory of action outlined above and advocated by the MI Excel staff (see Figure 4.1). However, the decisions made by district leaders created differences in the instructional expectations, communication/training plans, and teacher response to the instructional infrastructure system. Overall, the decisions made at the district and school level regarding ELA instruction were consistent with the administrators’ beliefs about influencing change at scale, whether taking an experimental science orientation or an improvement science orientation. 124 Overall, the mandated literacy program at Hillside relied primarily at designing a program that teachers with less experience and knowledge in literacy instruction could pick up and use to teach in a manner aligned with the district’s vision of high-quality reading instruction. This resulted in the adoption of the basal series Journeys along with a district-created curriculum that organized the textbook into units for teachers to deliver to their class. In contrast, at Lakeside, the administration mandated the literacy systems CAFE and Daily 5. Daily 5 provided a framework for how to organize the reading block into student-selected centers, while CAFE provided teachers with a menu of reading strategies that teachers should teach students as needed based on diagnoses from individual conferencing and small group interactions. This system relied heavily on a teacher’s ability to draw on her professional expertise to sequence learning opportunities, incorporate outside resources, and customize reading skill instruction for individual students and small groups. Fidelity Program I label the mandated literacy policy at Hillside as the “fidelity program” because the administration asked teachers to follow a new curriculum based on a basal series and curriculum guide. At Hillside, district leaders organized committees of teachers to write new reading and science curricula in the summer of 2016. The Hickory principal was a curriculum writer for the third-grade reading curriculum. According to her interview, there were two teachers from each grade, who worked to create curriculum units using the district basal series, Journeys, around a guiding question, such as “How can you be a successful person in society?” Journeys is a Houghton Mifflin Harcourt curriculum that the company described as “a comprehensive K-6 English language arts program” that addresses deep analysis of authentic texts, vocabulary, 125 spelling and grammar, writing skills, and listening and speaking skills (Houghton Mifflin Harcourt, 2018). In addition to the guiding question, these units defined weekly anchor texts from Journeys that connected to the guiding question; the reading, writing, and language skills that should be covered in the unit; word study topics; ways that the unit aligned to the district’s vision of HQI; and a performance task that every student would complete at the end of the unit (See Appendix Q for a copy of the 3rd grade curriculum framework for unit 2). Teachers were supposed to use these curriculum units in a framework of “purposeful literacy activities” and group work. Basically, teachers were supposed to give students “purposeful literacy activities” that they could engage with individually, such as working on a reading comprehension skill while independently reading, while the teacher worked with small groups of students in a guided reading format. Additionally, the district established a “master schedule” that defined the times each day that teachers should be teaching each subject area. Ms. O’Neill described teaching a week in this system: So that’s basically what it was, was reading basal stories for five days, for four days and then testing all day Friday because you had to give a vocabulary test, you had to give a comprehension test, you had to give a grammar test on things. Overall, the Hillside program asked teachers to teach reading to the curriculum guide that the district gave them. The curriculum guide denoted by bolding and highlighting concepts that would be assessed and that would count toward the data that would be collected and analyzed at district-level meetings. Teachers were also restricted to primarily drawing on the basal series as their resource to design the lessons within the curriculum units framework. 126 Origin of Plan. Ms. King, the superintendent described the problems in reading instruction prior to the 2016-2017 school year as: Last year, they were doing MASSA units but not really. And it looked like, felt like sort of reading and writing workshop without materials in their classrooms. So when I walked the classrooms when I got here, I saw a variety of materials, a variety of strategies none of which looked systematic across the rooms. Ms. King described the instructional expectations coming from work that Ms. Davis, the Hickory principal, did in her previous role as the instructional coordinator during the 2015-2016 school year as the source of what created the current instructional expectations. According to the superintendent, Ms. Davis surveyed teachers as to whether they would like to continue with reading and writing workshop, use a basal series, or try guided reading? Ms. King summarized the results, “Many of them said they wanted to go back to a traditional book.” Consequently, the district brought in different textbook vendors and ultimately the teachers chose Journeys. Ms. King outlined the positives of Journeys as: So we picked to purchase Journeys because one we have tons of new teachers, so building capacity of guided reading and writing workshop is not going to work. Two, we have class sizes of 30, 32 sometimes and that wasn’t working for us in the guided reading approach. And three as you are bringing in…having these units and having materials in their hands was really, really helpful. Basically, Ms. King expressed that she felt that it was more feasible to have teachers follow units and the Journeys system than to expect them to be able to implement a system that relied more on teacher knowledge, experience, or ability to work with small groups of students. 127 However, Ms. DeSutter and Ms. Lavin, experienced teachers at Hickory and Willow, expressed some regret with this decision, because at the same time as implementing this new curriculum the district hired reading instructional coaches. As Ms. DeSutter shared: We did not have that [the instructional coaches] last year, or ever before, and I think the joke is if we would’ve had that last year, we might have a different resource this year, because you got how tos … ‘Oh that’s how to do writer’s workshop. That’s how you do it!” Basically, they felt like with the coaches and additional training that district leadership had provided over the year, the workshop model could have been successful at scale throughout the district. The coaches could have been assisting teachers with developing the knowledge to teach to a less structured program. During the study year, they had the support, while teaching a program that may not need it as much. Connection to Administrators’ Experimental Science Beliefs. For both Ms. King and Ms. Davis, this plan reflected their experimental science assumptions in that the mandated instructional plan simply included teaching to a basal series and following the accompanying curriculum. The idea that teachers could learn the expectations for the next unit in a 2-hour meeting indicated that they believed that this curriculum was easy to “transport and implement” (Lewis, 2015) in different classrooms. Additionally, the required master schedule minimized variation between classrooms in the time of day and length of time dedicated to literacy instruction. They saw improvements happening in literacy instruction as teachers learned to better implement the curriculum. In fact, Ms. Davis shared that in regards to changes for next year’s literacy plan, the only change she planned to make was “just enforcing what was already written. Really making sure the curriculum is being followed.” 128 Following a more improvement science viewpoint, Ms. Nelson admitted to her staff that there would be variation in how teachers enacted the curriculum, and she used this variation to inform how she interacted with her teachers. In leading her teachers through the changing expectations for their literacy instruction, she said that she had told them the following: You might not agree with how that grade-level person is doing it, but you don’t have to. You have to teach what they’re setting out as the standards, and you have to fit it under this vision of high quality instruction, but that might look different in your classroom. She viewed and communicated to her staff that the curriculum was more of a guideline to inform teacher’s practice, but not as something that they needed to implement faithfully to the letter. This freedom allowed teachers to spend additional days on difficult content and allowed the second-grade teachers to teach it in a more reading workshop format, as will be described in more depth below. Professional Program I label the mandated literacy program at Lakeside as “the professional program” because the general framework of the program was to provide teachers with literacy systems to support high-quality practice but the program also retained a large amount of professional discretion in how the systems were used and when. During the 2016-2017 school year, the Lakeside School District focused staff on the Vision of HQ Reading Instruction rather than on a curriculum. District expectations for reading instruction were communicated in the form of the “High Quality Reading Instruction Implementation Guide” (in Appendix R). This document was readily available on the district website for all staff and community members to access. As described above as a part of the instructional infrastructure system, this document listed the teacher actions and student actions from the subject-specific vision of high-quality instruction with links to 129 further description and synonyms for each action. The Lakeside document then had links with actions that would be observable that would align with this vision. For example, at Lakeside teachers were required to teach using the CAFE and Daily 5 literacy methods. CAFE was a literacy program created by the 2 Sisters that provided teachers with a menu of strategies that reflect skills and strategies that proficient readers use. CAFE stood for “Comprehension, Accuracy, Fluency, and Expanding Vocabulary.” Each letter represented a different category of skills with strategies listed underneath. The CAFE program asked teachers to work individually with students to assess reading, explicitly teach reading strategies, and monitor student use of the strategy. Meanwhile teachers should keep records of student reading progress and use strategies to inform future conferences with that student. Then teachers should plan and facilitate small groups based on CAFE strategies rather than the typical student groups built around reading level. Whole class instruction was used to explicitly teach strategies that were needs for many students in the class. Lastly as a part of the program, teachers dedicated a bulletin board to be their CAFE Menu board (see Appendix S for an example). This board listed the strategies in each category and students posted their names next to the strategies that they were working on as a goal (Boushey & Moser, 2009). The CAFE book (Boushey & Moser, 2009) provided a strict structure for the first 10 days of school in order to set up the program, but after those 10 days, the system mainly relied on teachers using the framework to make decisions to best meet the needs of their individual students. As an additional resource, the district purchased the textbook series Making Meaning. As Mr. Dixon described, the textbook was brought in because they “thought that was a nice compliment to those programs and those approaches that would help teachers in teaching them strategies.” Making Meaning was created by the Center for the Collaborative Classroom and 130 provided texts that could be used to teach comprehension skills and expand vocabulary (Center for the Collaborative Classroom, 2018). No teacher used it as a basal series like was expected at Hillside. The Lakeside implementation guide provided a link with information on each of the expected strategies and programs which provides teachers with a description of the program, examples of observable behaviors, indicators that the strategy or program is working, and additional resources (See Appendix T for an example of the content in the CAFE link). Overall this implementation guide was the primary way that district leaders communicated their instructional expectations to teachers and instructional staff. Notable was that in the 2016-2017 school year, there was no approved curriculum in the sense of a pacing guide, instructional units, and unit-specific assessments. Basically, teachers were supposed to teach using instructional practices aligned with the visions of high-quality instruction using the CAFE and Daily 5 systems and Making Meaning as a resource. The Lakeside superintendent summarized the reading instructional expectations as the following: That’s our biggest thing like when you talk ELA, [it] is reading and making sure that we do best practices and we follow the systems that we say we’re going to do. In this case, guided reading, CAFE, using DRA strategies to drive instruction. We implemented Fountas and Pinnell as a part of our assessments this year and that’s been, that’s been very important for us as well as just having everyone a common language. In other words, during the 2016-2017 school year, teachers were to get familiar with using the instructional systems considered best practices by the district: CAFE, Daily 5, guided reading, and Fountais and Pinnell assessments. Actual curriculum defining the pacing and topics for each grade level were being developed during the study year. Consequently, teachers did not begin 131 teaching using this curriculum until the 2017-2018 school year and even when the superintendent described what the curriculum would be for the upcoming school year, he described, “It's not a new curriculum, what it is a realignment to units.” They planned for there to be unit topics that connected to social studies and science topics, but he stressed that it was not really a new curriculum. He explained, “It's really not. It's about taking the pieces we already had and making them more streamlined and more aligned with each other and make more sense out of what we had.” Origin of plan. When working to develop the plan for reading instruction, the Lakeside superintendent described seeking feedback from principals and teachers, using surveys, and looking at student data. District leadership also discussed with other districts what programs they were using. He described the teachers as being the driving force behind choosing the CAFE program: So you know, teachers enjoying and believing in what they’re teaching, feeling supported because of the training and resources provided to them, making sure then that it actually makes an impact, because you can bring in the most sexy program in the world, but if it’s not making a change in achievement then it is useless. In fact, he admitted that CAFE and guided reading were happening in pockets around the district for about 10 years, but rarely with fidelity. Consequently, the district’s plan became to train the teachers to better use the systems put into place in the past: We said “Okay, if we threw out the baby with the bath water type thing and we literally started over, what would we do?” And what we found ourselves saying is we’d probably go back to guided reading. We’d probably go back to CAFE. So, we weren’t wrong in some of things we tried to put together as a district years ago. We were just wrong in how 132 we implemented it and how we trained it and more importantly in how we supported it after training. That’s the part we were missing. The district then invested resources into initial training and embedded professional development, through training the principals to able to provide targeted feedback on the reading program instruction as instructional leader in their schools. This belief that the problem with their previous practices may be in the training was aligned with the improvement science assumption that programs need knowledge-building supports. The teachers at Cottonwood also acknowledged their role in bringing the CAFE systems and Daily 5 routines to their school and consequently their district. Ms. Johnson described her role as a co-leader last year in spreading these practices around Cottonwood: Last year we had a different principal and he actually let the teachers take a lead on what we thought was best practices for our building. So the kindergarten teacher was the one who spearheaded this and we all started doing, really, Daily 5 and we started doing the CAFE strategies. And the kindergarten teacher and myself, we actually went to Chicago and saw the Sisters [developers of CAFE] and brought back even more strategies and practices to use. As a result of bringing these systems into Cottonwood during the 2015-2016 school year, they saw drastic growth in student achievement scores, which caused district leaders to take notice. Regardless of this growth, the decision to take up and mandate districtwide instructional strategies of high-achieving teachers demonstrates a deference to the specialized knowledge of teachers in the field. 133 Connection to Administrators’ Improvement Science Beliefs. The plan was built based on trust in the people in the system. Mr. Dixon admitted that he had to go to the experts in his district because literacy instruction was not a topic in which he had extensive knowledge: Trusting the people. You know, I’m not a reading specialist. I was a band director. Trusting the people that know more about reading than I do and making sure the same exact thing. How do I give them all the support that they need? If they say, this is what we need to do in reading, I have them in a place for a purpose and for a reason. I believe in them, so I need to trust their judgement. If they ever need to learn to play saxophone, I hope they come to me. But if I need to learn how to make sure that our kids know how to open their book and read, then I need to make sure that I listen to people who know what the heck they’re talking about. And I give them everything I can to make sure that they can do their job effectively. This trust was shown by taking districtwide the practices of a few highly effective teachers at Cottonwood Elementary who had made the effort to research and get trained on the CAFE method. The structure of the literacy expectations also aligned with the improvement science beliefs. Instead of mandating a lockstep curriculum, district leaders mandated a system of instruction that gave teachers plenty of flexibility to use the systems to meet the needs of their specific students In fact, Ms. Wilson described a challenge in that the teachers tried to follow the district literacy instruction programs too rigidly: I think that because there were a lot of new initiatives, I mean whether it's Making Meaning or Eureka Math, I think the people just were so concerned with completing or doing this and not missing the details, they got a little bit too constrained. And you 134 probably saw that or even like a scheduling like oh I only have a literacy block from this time to this time, so instead of just using your professional judgment and say “you know I have to go half an hour over because these kids need it.” People tend to become very ritualistic and very routine-based and sometimes that stunts our growth. She hoped that during the 2017-2018 school year, teachers would be more willing to take the foundation that they built this year with using the instructional systems and use it more fluidly to meet the needs of their students. This demonstrated an appreciation for modification and variation in using district programs. Enactment Processes As has been extensively studied in the literature, it is not sufficient for administrators to simply provide new instructional programs, they must use levers to incentivize their staff to take up the program and alter their practice (McDonnell & Elmore, 1987). The levers that the Hillside and Lakeside administrators put in place also reflected the assumptions for change that they held. McDonnell and Elmore (1987) define four types of levers that administrators could use to translate policy goals into concrete actions: mandates, inducements, capacity-building, and systems-changing. Mandates refer to the rules that govern the actions of actors in an organization. Mandates represent required actions that all individuals are required to do. Inducements refer to incentives such as money in return for some sort of action. Capacity- building refers to the transfer of money in order to build future capacity, such as professional learning opportunities. Lastly, systems-changing is the transfer of authority in order to involve different institutions in the change process. In attempting to change teachers’ reading instructional practices, leaders at both Hillside and Lakeside used mandates, inducements, and capacity building to encourage teachers to make 135 the desired changes. This section explores how the superintendents and principals differently used these levers in alignment with their orientation toward change in an attempt to influence their teachers’ actions. Resources Curriculum. In Hillside, teachers were provided with a curriculum framework for each unit. Ms. Miller described the process of using the curriculum as: What we do is we go through the curriculum framework, and there’s stuff that is bolded and there’s stuff that is highlighted. So what we do is we go through and we pick out all the stuff that we have to do…the bolded is required. The highlighted is required and for data. The units in conjunction with the bolded and highlighted components ensured that all teachers were teaching the same topics and assessing in the same ways. Additionally, it ensures that all teachers were on the same schedule with their lessons. In Lakeside, teachers were provided with the Making Meaning textbook. However, unlike at Hillside, there was no pacing guide or units provided by the district that mandated how the textbook should be used. Instead, as explained above, teachers incorporated a plethora of different resources to support their use of the two instructional systems, CAFE and Daily 5. The Making Meaning textbook was simply an additional resource for the teachers to draw on when designing lessons. Additional Resources. In Hillside, teachers were also provided with the Journeys textbook and supplemental materials. While principals and teachers were told that the basal series, Journeys, was to be used only as a resource and that the curriculum framework was to be 136 used as the curriculum, there seemed to be confusion and fear regarding incorporating other resources into lessons for teachers at Hillside. Ms. Trainor describes her confusion as: If teachers are using something else, they might be hearing from their principals or other people, “You’re not allowed to use that. We have Journeys. This is what you’re supposed to use.”... but [if] you are telling me I can’t use this other resource, Journeys is what I need to use, then aren’t you forcing me to use Journeys as a curriculum?” She continued to describe this confusion as frustrating because at times she did not feel like she could do what is best for her students. This administrative practice of limiting access to outside resources further restricted teachers to the district mandated program. Even with these restrictions, teachers did mention at times incorporating outside resources such those found online. Ms. Morey mentioned searching on Pinterest when, “I have a question that they are struggling with, I am not afraid to look on the internet.” And Ms. Miller incorporated a resource from Teachers Pay Teachers into her small group time for her accelerated readers. Lastly, Ms. Trainor also mentioned using the The Reading Strategies Book: Your Everything Guide to Developing Skilled Readers (Serravallo, 2015) that she had access to from her continuing education focused on becoming National Board certified in reading. Overall, the use of outside resources was limited at Hillside to small supplements. Teachers at Lakeside had no fear of incorporating outside resources and listed a large variety of resources that the district provided them to support their use of the mandated literacy systems. Their uses of the resources were not as closely monitored as those at Hillside. Teachers mentioned using The CAFE Book: Engaging All Students in Daily Literacy Assessment and Instruction (Boushey & Moser, 2009), The Reading Strategies Book: Your Everything Guide to Developing Skilled Readers (Serravallo, 2015), Storyworks magazine (Scholastic Inc., 2017), 137 The Next Step Forward in Guided Reading (Richardson, 2016), the Lakeside implementation guide, and online resources such as Teachers Pay Teachers and Pinterest along with the Making Meaning curriculum in order to design units and lessons that matched the needs and skill levels of their students. Overall, the restrictions on outside resources at Hillside reflected the experimental science assumptions of Ms. King. By restricting resources, she endorsed the idea that the knowledge was in the program and that teachers simply needed to follow it. On the other hand, the ability of teachers at Lakeside to incorporate any outside ideas to assist their implementation of the systems reflected the improvement science assumption that there was also knowledge in the people and that variation in instructional resources could be a source of new ideas for future improvements. Training Professional Development. At Hillside, district leaders and curriculum writers communicated the curricular expectations to teachers primarily through half-day professional development days, known as “curriculum rollout meetings.” In these meetings, the curriculum writers would review the upcoming unit with their grade-level colleagues. Ms. Miller described these meetings: So we've had a half day where the writer of the curriculum, so it will just be second grade in the classroom, and the writer is one of the literacy coaches, just not ours, she rolls out the curriculum and like gives us a rundown, 'This is how we're scoring, this is what you need to score on, and this is the data you need to collect' and then we have time to like look through it and if we have any questions. So that's likes a huge support...the only thing is it's so far in advance, so like, I'm teaching unit three right now, a couple of weeks 138 ago we had roll out of unit four, and last week we had a roll out of unit five. So to ask questions, I'm like, I'm looking at it and I'm like “gotta go back and do this for unit three” For five of the six Hillside teachers who participated in this study as focus teachers, this was the extent of the professional development that they had access to (the last teacher, Ms. Trainor, was working on her National Board Certification in Literacy: English Language Arts and finishing her master’s in Elementary Education with a focus on reading instruction). Three of the teachers mentioned that they felt like these professional learning opportunities were insufficient to promote true learning. Ms. Lavin shared her frustration: Most of them was on, like I said the roll outs. So it's just, here's the unit and walking it through. But isn't really teaching like, how to teach it better, the best strategies....So sometimes I feel like it's, they throw things in just to give us information to make sure they have it covered, but it's not, it's not really, like put in five bullets in an email. Quiz me if you need to on those 5 bullets, but do I need to spend a half hour when I could possibly be digging deeper into my curriculum?" As Ms. Lavin said these professional development opportunities were primarily at the service of communicating expectations to teachers on how to adequately teach the mandated curriculum with fidelity. This district-provided professional development further reflected the experimental science assumptions that when implementing change the knowledge is in the program, that improvement occurs when the program is faithfully implemented and that the change should be easy to implement in this case that teachers can learn a whole reading unit in a half-day professional development meeting. Lakeside teachers had access to a variety of district-provided and outside-offered literacy-based professional development opportunities. The first way that leaders at Lakeside 139 planned to influence teacher practices was through before-school, districtwide training. For the new elements of the literacy program, such as the Making Meaning textbook, district leaders paid for the developers of the curriculum to come introduce the program to the staff at the beginning of the year. They also offered training before the school year began on the programs, such as CAFE and Daily 5, that district teachers had had some limited experience testing in their classrooms during the 2015-2016 school year. Additionally, many teachers indicated attending additional learning opportunities organized by their local Intermediate School District (ISD) on concepts like guided reading and read alouds with accountable talk. A handful of teachers from both Cottonwood and Rose also participated in a yearlong professional development program offered through the Intermediate School District (ISD) called “Reading with the Experts.” From this opportunity, they attended monthly meetings and received coaching from ISD reading experts through onsite observations and feedback. In fact, they were provided with so many opportunities to participate in further professional development that Ms. Olin complained that she felt like she had missed too many school days while in professional development. The difference between professional development opportunities between the two districts was that at Hillside, professional development was limited to learning focused on implementing the mandated curriculum, while at Lakeside teachers had the opportunity to engage in professional development that met their individual needs for learning around reading instruction. This demonstrated a trust in the teachers in identifying useful experiences and a belief that instruction improves as teachers have opportunities to learn how to teach and not simply what to teach, like at Hillside. Overall, Lakeside invested in developing the teaching capacity and knowledge in their teachers. There was a more professional orientation toward teachers becoming experts in their fields. 140 Instructional Coaches. At Hillside, teachers had full-time instructional coaches who came into classrooms, observed lessons, and provided feedback to teachers to support their use of the mandated curriculum. All 6 teachers described their coach as a huge support and resource for their reading instruction. In fact, Ms. Trainor described receiving coaching as one of the biggest positives from the school year, “I think the positive would be for sure having the coach. I thought that was super cool. That's the only person that I've been able to see teach to model for me, that'd be the biggest thing that was awesome.” Beyond providing support, both Ms. Miller and Ms. Lavin shared concerns about the instructional coaches’ ties to district leaders. Ms. Miller shared an instance in which an instructional coach simultaneously acted as an enforcer of the program: There was one time in November we had taken the skills from Journeys for writing and then we found a “Teachers Pay Teachers” unit with that same skill, because the one in Journeys was like, it wasn’t enough, our kids needed more guidance. So we printed it off and then we were at a grade level meeting for the district, and we talked about it, and we said that we would share it with them, whatever, and one of them told their literacy coach, different from ours, and we got central office called on us. Central office called my principal who then told us to trash it. The instructional coach served not simply as a source of support for teachers as they worked to teach with fidelity to district mandates, but also served as a compliance monitor by bringing information back up to central office leaders, who could further pressure compliance. In alignment with the experimental science paradigm, these coaches were another tool to monitor fidelity of implementation. 141 Districtwide at Lakeside, there were no full-time positions of instructional coach; however, at Cottonwood Elementary, Ms. Johnson served as an instructional coach in addition to teaching 2nd grade. In this role, Ms. Wilson provided her time during staff meetings to share what she was learning in her “Reading with the Experts” professional development that she was attending throughout the school year: She [Ms. Wilson] gives me time at staff meetings twice a month, usually I go once a month and I present. With the Jan Richardson, I made copies of all my notes and then at each grade level I copied information that would be important for them to use at the guided reading table at each level. I am trying to bring back and share as much as I can with the staff. Unlike at Hillside, Ms. Johnson could only engage in her role as an instructional coach during staff meetings, because she had her own classroom, which limited the time available for her to push into other classrooms. Even with this limitation, she used her role as instructional coach to attempt to further teachers’ literacy knowledge so that they could more intentionally use the systems in the CAFE and Daily 5, such as the guided reading time. Additionally, because she was embedded in Cottonwood and not in regular communications with central office leadership, she did not work as a district compliance monitor like those coaches in Hillside. It is worth noting that full-time instructional coaches began during the 2017-2018 school year and Ms. Johnson was hired as one for a different elementary school in the district. Mr. Dixon shared that at Lakeside a focus was to support teachers through embedded training throughout the school year. Consequently, at Lakeside, the goal was that BN members also were to serve as instructional coaches. Both Ms. Wilson and Ms. Reed mentioned that BN members drove the instruction in the school. Ms. Wilson described her BN members acting as 142 peer coaches. Ms. Reed described using her BN members to help her design professional development at staff meetings to address areas of instruction that the HQI data says are not being done. PSP Meetings Problem Solving Protocol (PSP) meetings were a driver in the Blueprint system. As described in the theory of action, PSPs were an opportunity for groups to analyze district data and create action plans to improve practice at the school level. At Hillside, PSPs occurred as a part of BN meetings. In these meetings, the BN would look at school data like survey results on teacher-student relationships and M-Step data. Then as a team, they would decide on action steps for the school. The clearest example was at Willow in which the BN analyzed survey results on teacher-student relationships and decided to create a plan with Paw-some notes to improve girls’ feeling that their teachers cared for them (this example was described in more detail above). At Lakeside, PSP meetings were conceptualized as a form of embedded training. At Lakeside, PSPs occurred monthly and all staff sat in the same room at tables with their grade level colleagues to analyze grade level data, including NWEA scores, Fountas and Pinnell reading levels, and HQI walkthrough data. In their grade level teams, teachers then discussed ways to address weaknesses in student performance. As described above, the routines for these meetings were being refined throughout the school year. By the May PSP meeting, teachers at Cottonwood decided that they had finally found an effective form for their data analysis processes (see appendix U). As you can see in Appendix U, the PSP form required teachers to plan for interventions to meet the reading level of each students based on the data that they analyzed and then teachers would check in with their grade level colleagues the following month to see how the plan worked. 143 At both districts, PSP meetings involved teachers in the plan-do-study-act cycles common in improvement science paradigm for scaling up change. The difference was in who had access to the data and the potential of PDSA cycles to lead to refining of the literacy instruction in classrooms. At Hillside, the plan-do-study-act cycles only officially occurred in the BN meetings. Monitoring HQI walkthroughs. As described in the previous chapter, HQI walkthroughs were a component of the instructional infrastructure system in the Blueprint program. HQI walkthroughs asked school principals and other district leaders to complete walkthroughs of classrooms to observe instructional program implementation and collect data on teacher practices. While this system was common in both districts, the practices of principals in each district differed in ways that related to different perceptions from teachers on the practice. The major difference between the districts was around the communication about and throughout the system. At Hillside, both principals and central office leaders made an effort to engage in HQI walkthroughs. Principals were under the impression that the purpose of HQI walkthroughs was primarily for the collection of data on teacher practices. This impression caused them to not provide information to teachers on the purpose of the walkthroughs as well as to not provide feedback so as to not skew the data at future observations. At a December 7th meeting, it was clear that they were under this impression from some prior interaction with the superintendent. At that meeting, Ms. Davis asked Ms. King if she could share the observation form with her teachers because they had been asking to see it. Ms. King agreed and as the year went on, principals began sharing the observation form and providing feedback to their teachers. However even at the end of the year, all six teachers in Hillside indicated that they had to ask 144 their principal for feedback and that they continued to receive no feedback from the superintendent and other central office leaders when they completed their observations. These decisions to not share the purpose of the HQI walkthroughs with teachers made them less sure about the purpose of the walkthroughs and whether their teaching was meeting expectations. When the teachers were asked the purpose of the HQI walkthroughs, all the teachers at Hickory Elementary knew that the principal was looking for specific practices. Ms. Morey described that, “Usually when our principal comes in, we know what she is looking for. It’s like a specific topic or she has passed out papers to us.” As Ms. Morey indicated her principal, Ms. Davis made a choice at their first staff meeting in January to share more information about the HQI walkthroughs. On the other hand, teachers at Willow were not as clear about the purpose, because Ms. Nelson did not make time for a similar staff meeting. Ms. Trainor knew about the “list of attributes they’re looking for”, but Ms. Lavin shared that they did not know and “I mean, we would assume that what they’re looking for is all the parts of the Danielson. They’re looking to see if you are…if the students are engaged” Additionally, all six Hillside teachers expressed feeling stressed, uncomfortable, or under pressure because of these HQI walkthroughs. They all credited the lack of feedback as contributing to the stress associated with the observations. Ms. Trainor shared that while at her former districts she was used to being observed regularly, there was a difference in how it felt at Hillside. She described: The fact of them [the principal and central office staff] coming in does nothing. It's actually after they leave that makes me feel a little more stressed out because they leave and there's no necessarily acknowledgement of anything. That's very stressful for me. 145 The lack of feedback left teachers feeling like they “always have to be on your toes.” (Ms. Lavin), on schedule with the master schedule, and teaching to the curriculum. Overall, the HQI walkthroughs felt to these teachers like a mechanism to ensure their compliance with the literacy systems. Ms. Miller shared her perceptions of the purpose of the walkthroughs: When Ms. Davis comes in, and she is just popping in, and sitting there, she is observing you for several things. One is for data, which is she has a form and she's looking for certain like buzz things that are showing ambitious teaching. That goes into like a database of the whole district, and as principals they analyze it. It's not really connected to us in any way. But, since she's the principal and she does our evals, she also can take good things or bad things that she sees and uses it towards our final eval. It was clear that Ms. Miller recognized that her principal is collecting data to measure her teaching; however, she felt like it is not really “connected” to her. While Hillside teachers at times did receive some individual feedback, another difference was that teachers at Hillside did not have access to the school-level data. No Hillside teachers mentioned hearing anything about their school-level or district-level HQI data in any of their interviews. In fact, Ms. Morey discussed how over the year, she knew that the administration team was looking at the data, but that she did not believe that the teachers would get any information from the data until the next school year: They are trying to collect data. It’s my understanding that there are no names attached to it, but it’s attached to like a grade level and maybe a building and seeing like what practices we’re using and what practices we’re not using so that, in order to be highly effective, everyone could be following the same practices, so I think it’s something that maybe they’re just taking data this year, analyzing it and then maybe next year they’re 146 going to be like, this is what we have seen, this is what we need to fix and all that. So yeah, I’m always observed all the time and it’s just, you never know. Ms. Miller and Ms. Morey shared their perceptions of these HQI walkthroughs being used as measurement of their instructional practice but not a source of individual or school-level feedback. This was consistent with experimental science assumptions in that the data served as an implementation measurement tool but was not used in a way that led to knowledge building about her practice through plan-do-study-act cycles like what would be found in improvement science. Like in Hillside, at Lakeside, principals implemented walkthroughs during the 2016-2017 school year. Throughout the entire school year, some sort of feedback to teachers was a part of their system. This practice resulted in teachers knowing from the first day the purpose of the walkthroughs in terms of the instructional practices that their administrators were looking for during their time in the classroom. In fact, when asked about the purpose of the walkthroughs, all six teachers could accurately describe the reason their principals were visiting their classrooms, the criteria on the data rubric they were collecting, and their goals for their instructional practice. Aligned with improvement science practices, the system for providing feedback evolved as the year progressed and as principals learned better systems for providing meaningful feedback in ways that did not require a large amount of their time. Additionally, as described above Ms. Wilson and Ms. Reed individually did more to support their teachers’ understanding of the purpose of the walkthroughs. Ms. Wilson hosted a reading night for her teachers to further learn about the reading vision of high-quality instruction and to play the role of the administrator collecting data on a videotaped lesson. Ms. Reed facilitated opportunities for her teachers to observe their colleagues and complete the HQI 147 walkthrough data form so that her teachers also could have some experience in looking for the practices as well as more opportunities for teachers to receive feedback on their instruction. Lastly, administrators at Lakeside shared the data from the HQI walkthroughs not only with their BNs but also with their entire staff at PSP meetings. This sharing of the data allowed every teacher to see how the data was being used and how it looked in an aggregated format. At the PSPs where the administrators shared the data from HQI walkthroughs, the staff set goals for their instructional practice based on what practices were less evident in the practices of teachers at the entire school. In both Cottonwood and Rose, this resulted in teachers focusing on ways to incorporate more text-to-text connections. In April, I observed Ms. Stewart teach a lesson in which she did read alouds of two fables from two different cultures and students worked in small groups to create Venn diagrams to compare and contrast the two texts. When asked how she chose to teach that target, Ms. Stewart shared: When we looked at the high-quality data that Ms. Reed collects when she does our classroom walkthroughs, it was evident in the whole school that no one was really hitting that standard, or if they were, she wasn't walking in and observing it. She asked us to intentionally plan a lesson on that. This was that lesson. This lesson and Ms. Stewart’s rationale provided evidence of the data being used to influence teacher practice. Ms. Edmund and Ms. Clark also mentioned focusing on this skill in their instruction due to the HQI walkthrough data. This overall transparency around HQI walkthroughs contributed to the teachers’ comfort in the observation process. In fact, Ms. Johnson actually described these walkthroughs as being one of the best parts of the 2016-2017 school year: 148 I think Ms. Wilson’s done a great job this year with walkthroughs when she comes through she gives feedback, “This is what I saw you do. What do you think about this?” And I’m telling you, it made me think a lot more about my lessons and how to make them more beneficial for everyone, you know. I really took what she said, coming in with a fresh set of eyes in what could be worked on. And I think that, I think that made me a better educator this year, I do, I think it helped me. It pushed me to be more thorough, really hit different levels of questioning, not just surface, I mean, really understand. That’s why I went to inferencing. Let’s try to get deeper. While not every teacher shared Ms. Johnson’s enthusiasm for the walkthroughs, only two teachers at Lakeside described being stressed by the process, Ms. Edmund and Ms. Stewart. Notably, Ms. Edmund was the only teacher at Cottonwood that did not attend the reading night that Ms. Wilson hosted to explore the reading vision of high-quality instruction. This missing information may have contributed to her stress. Ms. Stewart on the other hand, shared that she got stressed when the principal entered her classroom and that she responded by, “Sometimes when she [Ms. Reed] walks in, I'll go, ‘observation, data walk, visiting?’ and she'll say, ‘Oh data walk,’ and I'm like, ‘Oh, okay fine.’ If she'll say observation then I'll go, ‘Ghuahh’.” This quote made me believe that at least to Ms. Stewart the HQI walkthroughs (data walks) might not be the source of her stress at seeing her principal, unlike her teacher evaluation observations. In fact, she shared that in the end, these walkthroughs have been “more reassurance because you know you do so many of these things, not intentionally. If she stays in the room long enough, she's going to see almost all of them, so I guess it hasn't really affected me.” Overall, the choices made at Hillside made HQI walkthroughs more of a monitoring system like would be found in an experimental science system. At Lakeside, using the data in 149 PSPs allowed the HQI walkthroughs to be part of a system of plan-do-study-act cycles in which teachers can reflect on the practices that are not being addressed and are able to make action plans for focusing on incorporating those standards in the future, such as when they concentrated on incorporating more opportunities for students to engage in text-to-text connections. Physical look-fors. Teachers at both Hillside and Lakeside school districts expressed that they thought that there were physical classroom items that their administrators expected to see when they entered their classrooms. At Hillside, the district expected to see anchor charts posted around the room on lesson topics from the year (see Appendix V for an example). While Ms. Lavin did not agree with the practice of anchor charts, because she had a preexisting instructional practice of students practically making anchor charts in their own notebooks, she incorporated posted anchor charts to meet this district requirement. She reflected, “That’s where they call us out on if they’re not posted up and even in how nice they are.” Ms. Lavin also described feeling pressured to use the large sticky notes like the chart in Appendix V. At Lakeside, teachers indicated that administrators were looking for a CAFE menu bulletin board (Boushey & Moser, 2009) (see Appendix S for an example). Ms. Olin shared what administrators were looking for: When administrators walk in, they want to see that on, on the board for all, all, we have letters given from the district, but it’s comprehension, accuracy, fluency, and expanding vocabulary. Which when working with the CAFE, the Daily 5, those are the strategies that are supposed to build better, stronger readers. So that’s where that comes from. The CAFE Board was a simple, visible indicator of a teacher’s use of the CAFÉ System. Overall, this practice of administrators monitoring for simple, physical indications of following the 150 district expectations may be a sign of some experimental science assumptions in the Lakeside implementation processes. Master Schedule. At Hillside school district, every teacher was also given a “master schedule.” The master schedule dictated at what time every day they should be teaching each subject. This master schedule was a commonly brought-up topic in interviews by the participating teachers. Four of the six teachers that I observed and interviewed at Hillside communicated that teaching to the master schedule was a challenge; however, there was immense pressure to stay on the schedule. Ms. DeSutter mentioned in her interview that she had heard that, “not here, at other schools they [administrators completing walkthroughs] had made complaints, that I came in to observe science and you're teaching something else.” Consequently, on top of the high-quality instruction teaching practices, teachers were always conscious of staying on schedule. Additionally, this master schedule further constrained teachers’ practice to following the district program reflecting the experimental science assumption that variation, even in lesson timing, is problematic. At Lakeside there was no master schedule; however teachers did have specials time and a set time when instructional aides were in their classroom to support guided reading. This system allowed for more teacher decision-making in their classroom regarding the time and length of each instructional block. Allowing for more teacher discretion in scheduling made it possible for some teachers to have multiple reading instruction times throughout the school day and others to dedicate the entire morning to reading. Allowing for this type of variation in how reading was structured reflected the improvement science orientation toward bringing instructional strategies to scale. 151 Data collection. Like the walkthroughs, incorporating data collection systems on student achievement was a component of the Blueprint; however the particulars of the system were unique to each district. At Hillside, once a month teachers were required to submit the number of students in their class that scored above 90% on the unit assessment and on the unit checklist. Ms. Lavin described the process as “If they don’t have a checkmark, because they have to have 90% in every section, they can’t get 90% and say that they are passing, proficient.” These performance tasks and unit assessments referred to texts in the basal series, which further encouraged teachers to teach those anchor texts in order for their students to perform well on the assessments reported back to administration. At Lakeside, teachers were to enter students’ DRA reading level monthly according to their running records assessment as well as the CAFE strategy that each student was working on at that time. Ms. Johnson described the system as “I have to put the level, their text level, and I have to say what strategies we’re working on” into an online system called “Pensieve.” This data collection system required teachers to complete running records assessments with each student, which typically occurred during individual conferencing, a required component of the Lakeside instructional framework. Additionally, entering a CAFE strategy also required teachers to be using the system with students. While this data collection system was mandated, it could also be used to inform teacher practices. As Ms. Stewart explained the reporting system, “will let me track and see which strategies they’ve had, how long they’ve had it for.” Teachers daily logged into Pensieve while conferencing with students making the data collection a part of their instructional practices rather than an add-on reporting system. Some teachers reported using their data in the Pensieve system to study their own data in order to improve their practice. Ms. Stewart reported: 152 We use the data from our pensive, our conferring notes, to place kids into reading groups, and so like hit the CAFE strategies and stuff like that. The kids are using data in their binders, so they know where they're at most of the time, or they should know where they're at. Yeah, I mean data's a daily thing. Others shared that the data points were still too broad to be helpful to inform practice. By simply reporting the reading level and skill, the data was not targeted enough to inform instructional practices. Ms. Clark shared that: I love the idea, but I don't think that we're to the point where we can use that to drive instruction yet. Actually, I know that we're not there. The data that we're collecting is not specific enough. … We don't have the information to really pinpoint it into smaller focus areas. While the data collected by Lakeside teachers for the performance management system was more authentic to the classroom, only some of the teachers were able to use it in their own practice to inform classroom instruction. Some Lakeside teachers also reported PSP meetings as an opportunity to examine the data that they collected on student reading achievement to inform their instructional strategies. As described above all of the data reported to the district level was also used by the teachers in plan-do-study-act cycles through their PSP meetings. Ms. Johnson described the process that she and her second-grade colleagues went through at a May PSP meeting to analyze their classroom data and inform their instruction: She [Ms. Wilson] asked us to take our NWEA scores, we had to take the kids and we had to put them into three piles, red, yellow, and green. And then, and what [colleague] and I did is we thought okay these reds are all below this level, yellows are all here. And then 153 we looked at the NWEA and it says ready to learn, what they needed. And then I looked at that reading coaching, Coaching for Readers book. And what I did, what we did was then we looked at level, like, say, um, levels 20 to 24. We found common strategies in those two levels that the kids, we thought the kids needed to move ahead….so then we meet in two weeks, we’re gonna say did we do this, did it work, and what are we gonna do next? At Lakeside, PSP meetings were an opportunity for all teachers to engage in PDSA cycles with their grade level data to refine their reading instructional practices using the CAFE model. The district’s enactment of the PSP system using reading data was consistent with their improvement science orientation. Lesson Plans. Only at Hillside School District were teachers required to submit lesson plans to their administrators following a district-provided template, which aligned with the district’s curriculum. Ms. Lavin described them as “the way that it’s laid out, it’s really just to prove to them that we planned and that we kind of have the big idea behind it.” Five of the six teachers at Hillside mentioned submitting these lesson plans to their administrators. The teachers at Willow said that this was a required practice. At Hickory, Ms. Morey said that she and her teaching partner had shared their Google Doc with their plans with the principal, while Ms. Miller did not feel like it was required to share her lesson plans with her administrator, but that they needed to be prepared, because “if she comes in and she feels that things weren't going well she'll ask to pull them.” None of these teachers said that they received any sort of feedback from their administration on their submitted lesson plans after their original submissions of the school year. Ms. Trainor shared: 154 I haven't gotten any feedback this year after the initial submissions. So in the beginning of the year when I submitted my lesson plan, I did get feedback. I use the same template over and over again, so I haven't really gotten any feedback since then. Probably because I assume it just means I'm doing what I'm supposed to be doing. The feedback that she got originally was around missing pieces in her lesson plan from the district lesson plan template, which demonstrated again a focus on compliance to the systems mandated by the administration and the need for principals to be monitoring implementation. Monitoring with Flexibility Multiple teachers at Lakeside mentioned that their interactions with their administration had permitted them to exercise some professional discretion in using the mandated system. First, Ms. Stewart mentioned that she was not quite teaching to district expectations at the time as far as how she was grouping her students; however, after she explained her rationale, her practices seemed to be allowed by her administrators: Whenever we meet with our district personnel, they always say, “Oh how are strategy groups going?” I always say every time, “Well, it’s hard because every time I meet with a kid, their need is not the same. They might be reading a different text or their mindset might be different, or-“ and they said, “That’s okay.”… It seems like that’s been accepted. In a more drastic example, Ms. Parrish met with district personnel at the beginning year about the possibility of teaching reading strategies through social studies and science materials. She described the process as “we had asked their permission, Mr. [the Director of Instruction and Evaluation] and Ms. [the Assistant Superintendent] at the beginning of the year if we could do 155 it.” Then at the end of the year, Ms. Parrish and her teaching partner made a presentation to the school board and central office administration. Ms. Parrish described the presentation as: We did a whole presentation, a PowerPoint with what strategies were, you know it was kind of one of those things Kim, it was wonderful, because when we got done, I don’t think we realized Mr. [Director of Instruction and Evaluation] had said, you know, he said, “I counted 14 strategies right here, just in what, you know, in what you were talking about.” And I thought, “Oh my gosh. I didn’t even realize.” As described above, based on the work of this teacher team, the district’s curriculum framework for the 2017-2018 school year based its reading units on social studies and science topics. While this second case may indicate the experimental science assumption that customizations should be approved, the not only acceptance but support for the customization better reflected the improvement science assumption that variation can be a source of ideas to improve practice. In fact, based on Ms. Parrish’s success, the practice of teaching reading through social studies and science materials was incorporated into the upcoming Lakeside curriculum for the 2017-2018 school year. These examples of administrative monitoring while allowing for professional flexibility further demonstrated the improvement science assumptions of change in that variation was allowed and supported in both Cottonwood and Rose throughout the 2016-2017 school year. Summary Overall, these implementation strategies reflected the theory-of-change assumptions of each administrative team. In Figure 5.1, I present a summary of each district’s implementation strategies and their corresponding paradigm. 156 Experimental Science Paradigm Nature of Scale Up Hillside School District Lakeside School District Improvement Science Paradigm Hillside School District Lakeside School District Monitor implementation fidelity at new sites Design or approve customization of program if needed Use incentives to implementation if achieve needed - Instructional Coaches as Monitors - HQI Walkthroughs without Feedback -Physical Look-Fors include Anchor Charts -Lesson Plans Collected Expect modifications as program is integrated with local knowledge systems. -HQI Walkthroughs with Feedback -Physical Look- Fors include CAFÉ Board Key activities include building consensus on the importance of the reform and how to measure its impact Key activities include mapping the drivers of the reform in the current system and how the program would change those; understanding variability in sense-making Key activities include using rapid PDSA cycles to enact and study program elements, refining them as warranted -Grouping Students by Reading Level -CAFE during Social Studies & Science Instruction - Ms. Wilson’s Reading Night - Cottonwood’s BN revising PSP documents to meet data analysis needs -Ms. Wilson’s Reading Study -PSP Meeting with BN members only -PSP meetings with full staff Table 5.1:Experimental Science and Improvement Science Orientation of District Literacy Program Implementation Strategies (as defined by Lewis, 2015, p. 56) 157 Table 5.1 (cont’d) Assumptions Knowledge is “in” the program -District-provided professional development with CAFE and Daily 5 training - District-mandated lockstep curriculum and basal series, Journeys - District-provided professional development, “curriculum roll-outs” Knowledge is also “in” people and the systems that use that program - Teachers limited in their ability to incorporate resources outside of Journeys into their lessons Program may need modification, driven by ongoing practical measurement Improvement occurs through faithful program implementation Table 5.1 (cont’d) 158 -District-mandated literacy programs, CAFE and Daily 5, which allowed for professional knowledge - Teachers allowed to choose and incorporate outside resources into lessons. - Teachers led some district professional development sessions on CAFÉ - District supported teachers attending outside professional development - District leaders regularly modified the HQI walkthrough form Variation can be source of ideas to improve program and site Variation is problematic Measurement - District-mandated program limited variation in instructional content, practices, and pace. - Teachers were limited in their ability to incorporate resources outside of Journeys into their lessons. - Teachers were required to teach to the master schedule Use well-validated tools to measure implementation and impact - Used NWEA data - Used NWEA data tools to test leading indicators predicted by theory of change; use Use practical measurement “balancing” measures to check for adverse impacts - Used numbers of students who scored above 90% on performance tasks and unit tests. - Administration encouraged teachers to incorporate outside resources – Administration allowed the 5th grade team at Cottonwood to teach CAFE strategies using social studies and science content. - At PSP Meetings teachers analyzed reading data to alter their precious practice and plan for future instructional strategies. - Used data on the strategy each student was working on and their reading level. 159 Table 5.1 (cont’d) Optimal Improvement Conditions Sites: Success most likely if new sites are similar to sites where program was proven Program/Tool: Easy to transport and implement, foolproof - Ms. King mentioned that they chose Journeys because it would be easy for new teachers to learn. Sites: New sites need not be similar; success depends on organizational knowledge- building systems as well as program fit Program/Tool: knowledge- building motivation, ownership, and customization in many settings - CAFE and Daily 5 transferred from Cottonwood to other Lakeside schools with support from training. - District leaders regularly modified the HQI walkthrough form -Ms. Wilson’s Reading Study - Cottonwood’s BN revising PSP documents to better serve their needs 160 As illustrated in Table 5.1, district leaders at Hillside and Lakeside made different decisions when implementing instructional infrastructure components into their schools and classrooms. While both Hillside and Lakeside did have some implementation actions that fell in each paradigm, on the whole more Lakeside actions followed an improvement science orientation toward scaling up high-quality literacy instruction, while more Hillside actions were aligned to the experimental science approach. A quick look at Table 5.1 shows that the majority of implementation activities at Lakeside were aligned with the improvement science paradigm, 13 compared to three aligned with the experimental paradigm These implementation strategies at Lakeside prioritized building teachers’ literacy instructional practice through access to outside resources and involving all teachers in plan-study-do-act cycles so that teachers could continue to refine their practices and administration could refine their systems. While at Hillside only two implementation practices were aligned with the improvement science paradigm and their remaining 12 actions corresponded to the experimental science paradigm. Overall, the implementation strategies at Hillside prioritized training at the service of implementing the program faithfully and monitoring teacher implementation of the program. The clearest contrast was between the data collection and usage between the two districts. PSP meetings occurred in both districts allowing staff to engage in a form of PDSA cycles, a central component of improvement science. Additionally, in both districts some PSP meetings analyzed NWEA data demonstrating a deference to “well-validated tools” to measure instruction, a principle of experimental science. Importantly, PSP meetings included different staff members allowing for PDSA cycles to be used for different purposes. At Hillside, all of the data collected, whether it be from HQI walkthroughs or student achievement, were used by administration to monitor teacher progress in the system. In contrast, at Lakeside, while the same data was 161 collected and used by administration, this data was also used by all teachers in the district at staff meetings and PSP meetings as discussion topics and to inform future practice, such as through focusing on incorporating more instruction on the reading skill of making text-to-text connections. These leadership assumptions, program choices, and implementation strategies influenced both teacher beliefs and implementation of the mandated systems. The next section explores the implementation practices of the 12 participating teachers. Teacher Implementation of District Literacy Expectations After all of the administrative decisions and implementation systems were installed, it was up to the teachers to implement the district’s reading expectations into their classrooms. Like with the administrative practices, I noticed differences between districts in how teachers responded to the district’s mandates. Hillside Implementation At Hillside, teachers’ implementation and orientations toward the reform seemed to be based mostly on individual backgrounds and beliefs. Teachers could be classified into the following categories: Faithful Implementers, Adapted Implementers, and Resisters. Faithful Implementers – “This to me is normal” (Ms. Miller). Two first-year teachers participated in this study, Ms. Morey and Ms. Miller. As new teachers, they were trying to follow the district’s expectations as much as possible. While they both mentioned occasionally using internet resources, they both listed the Hillside curriculum, Journeys, and curriculum roll- outs as their biggest resources when designing their lessons. In their lessons, both teachers remained consistent throughout the year in teaching to the Hillside mandated curriculum and 162 reading block structure. Ms. Miller talked through her process for preparing for lessons as strictly following the Hillside structure: What we do is we go through the curriculum framework, and there's stuff that is bolded, and there's stuff that is highlighted, so what we do is we go through and we pick out all the stuff that we have to do…..The bolded is required. The highlighted is required and for data. Meaning like you have to collect it and grade it for data. And then we keep them, and we bring like they'll ask for a high, middle, low, and they want to see how it looks. That's what we do first. And Mondays, this particular lesson to plan for was pretty easy because you know you have to introduce the vocab words before the story, and nothing really in today is required for data other than the questions. So we are required to have an anchor chart for the question, and required to have an anchor chart for whatever the skill is. As first year teachers, they mainly spent the year trying to understand and follow the frameworks and expectations. They followed the Hillside structure guidelines for their ELA block, for example, Ms. Morey’s block followed the same structure throughout the entire school year: phonics, shared reading, handwriting, guided reading, purposeful literacy activities, and silent reading. The sole exception was when she joined with the other kindergarten class to do buddy reading during my last observation, which was a routine for Fridays. Ms. Miller’s block followed a similar format adjusted for the 2nd grade level. Early in the year, Ms. Miller and her grade-level team tried to incorporate outside material purchased from Teachers Pay Teachers to supplement the district’s curriculum and it resulted in central office leaders contacting her principal. However she explained that this was not an intentional skirting of the expectations, but a misunderstanding: 163 We were like, well, we didn't mean to be like sneaking this unit. We shared it at our district meeting. So stuff like that, they really just don't want. There was a lot of back and forth because they say you can pull certain things, and then they don't exactly know, I don't think. They didn't want us to be pulling in a completely new unit. One or two things, they felt we were pushing it, I guess, which I can understand, but we didn't know, but once we knew, we never did it again. This lack of intention of breaking with district expectations made Ms. Miller’s practice different than the adapted implementers and resisters. She was trying to supplement the mandated curriculum with what she felt were deeper tools. However, when she was corrected by central office leaders and her principal, she quickly returned to implementing faithfully the district curriculum for the rest of the school year. Figure 5.1: Hickory Elementary Literacy Advice Network Figures, 5.1, 5.2, 5.3, and 5.4 were created using the Kliquefinder software (Frank, 1995) to visually represent the informal subgroups around ELA instruction at each focus school. In 164 each figure, staff with the box symbol indicates members of the BN. The different colors represent different subgroups. Teachers participating in observations and interviews are labeled with their pseudonym. The size of each nodes represents that teacher’s usage of aligned reading instructional techniques as indicated by their responses on the fall survey. Besides the curriculum materials and district mandates, both teachers indicated that their grade level colleagues were another big influence on their instruction. As is evident in Figure 5.1, both Ms. Morey and Ms. Miller held close positions to their grade level colleagues in the Hickory Elementary School literacy advice network. Both Ms. Morey and Ms. Miller planned collaboratively with their grade level teams. Both the kindergarten and second grade teams seem to have higher fall aligned teaching practices based on the sizes of their nodes when compared to fifth grade. Additionally, their collaboration brought some other perspectives into their opinions of the mandated curriculum and the Blueprint overall. As Ms. Miller explained: A lot of people who have been here many years had time to really be mad about the Blueprint and all the change, whereas I didn't know what they didn't have before. This to me, is normal. But there's also a lot of great things like just last year they didn't have a social worker in their schools. I can't even believe that. I can't. I mean, there's gotta be, there's good things coming from it. I don’t know, I 'm not completely, I don't feel horrible about it. I mean, we'll see. I hope it works out. I do. Overall, these first-year teachers represented the truest attempt at implementing the district mandates of the curriculum and Journeys text of the teachers that I observed. Throughout all of their observations they taught according to the expected schedule and continued to rely on materials from Journeys. That being said, now that they have a year of experience both teachers expressed a desire to make some modifications to how they taught in the upcoming year. Ms. 165 Morey shares that over the summer she plans to reflect on "What can I do differently that will still meet the target but approach it in a different way?" This could indicate that during the 2017- 2018 school year, they may transition more into the “adapted implementers” category. Adapted Implementers – “Trying to adapt a direct instruction lesson into a mini- lesson format." (Ms. Trainor). Adapted implementers were teachers who intentionally implemented portions of the mandated curriculum in a way that worked with their beliefs of what good teaching was and their pre-existing frameworks. Ms. Trainor and Ms. DeSutter worked collaboratively as a part of the second-grade team at Willow Elementary School. Both are experienced teachers with at least 9 years of experience. However, this was Ms. Trainor’s first year at Hillside. During the 2016-2017 school year, Ms. Trainor was finishing up her National Board Certification in Literacy. The deep knowledge required for achieving national boards made her a critical consumer of the Hillside curriculum. She also wrote the lesson plans for reading for her entire grade-level team to use. Her planning process always began with the mandated curriculum: I always start with the curriculum guide, the Hillside curriculum guide. That guide gave me the book that I used. It gave me the teaching standards that I wanted to teach to, not specific to that book, but it gives you a menu of what we should be covering throughout the whole unit. From this starting point, Ms. Trainor used her background knowledge in literacy to incorporate these learning targets and Journeys materials into the lesson structure that she believed was most beneficial to student learning. In adapting the curriculum, she cited referring to a wide range of materials when preparing lessons. She mentioned using the required vision of high-quality instruction, the Hillside curriculum, and Journeys. But she also shared that she relied heavily on 166 her personal background and beliefs around effective literacy instruction. She stated, “My lesson looked a little bit different than what Journeys would have them do because the way Journeys is set up they don’t start writing until the third day. I try to provide them time to write every day.” She did not do the “purposeful learning activities” that the curriculum framework recommended. Instead, her classroom structure involved a whole class mini-lesson followed by 30 minutes of independent reading and reading groups. When teaching, Ms. Trainor also relied on her extensive reading instruction knowledge and resources to further meet student needs. During her May 22nd lesson, she was referencing a binder when individually conferencing with students. When asked about it, she explained: Probably not supposed to use it, but I’ll show you. But it’s from a different curriculum or a different resource, but it is a learning progression for students when narrative reading, so like I particularly looked at it with [student’s name] when I was listening to him read and talking with him and it was kind of helping me troubleshoot like exactly what would be a good learning, like what would be a good goal for him As the grade level’s planner of reading instruction, her plans that used the Journeys and Hillside curriculum in her adjusted structure became a standard across all three teachers in her grade level. Ms. DeSutter was the most experienced teacher in her grade level. However, she shared that Ms. Trainor had a deeper understanding of literacy instruction and she was happy to follow her plans. She regularly asked Ms. Trainor questions about teaching to the curriculum, “You do have a learning curve this whole entire year because I'm still saying to Ms. Trainor. 'How are we doing this?' or ‘What does this look like?'” In following Ms. Trainor’s guidance, Ms. DeSutter was also adaptively implementing the required curriculum. 167 Figure 5.2: Willow Elementary Literacy Advice Network As shown in Figure 5.2, Ms. DeSutter and Ms. Trainor’s positions in the literacy advice network at Willow Elementary are indicated by their names. It is interesting to note their central location and their proximity to both the principal and the literacy coach. This central position may have helped them to feel more supported in both implementing the program as well as in making the adjustments that Ms. Trainor felt beneficial. Additionally, Ms. Nelson, their principal was the one administrator at Hillside who communicated more improvement science assumptions about change. This type of leadership may also be contributing to their ability to adjust lesson content into the reading workshop format. Resisters – “I have kind of gone rogue” (Ms. O’Neill). After initially attempting to teach to the mandated curriculum and Journeys resource, by the end of the year both Ms. Lavin and Ms. O’Neill were intentionally not following the required materials, except what was assessed in the performance tasks. It is important to note that the structure of Ms. Lavin and Ms. O’Neill’s lessons tended to look very similar to those of Ms. Trainor and Ms. DeSutter; however they warrant a different category due to the outright resistance to the reforms that they expressed in their interviews. Interestingly, Ms. O’Neill was a curriculum writer for the 5th grade curriculum. However, she still did not follow the curriculum expectations. 168 They both decided to revert back to their former teaching practices, while being informed by the new program. Ms. O’Neill described it as: So then I’m using it as the resource but it’s not my primary resource anymore…. So planning is pretty much, I’m kind of going back to what I did last year, Readers Workshop. That’s what the instruction grammar calls for is workshop with guided reading added in that we didn’t do, at least in the Upper El last year or anytime. That’s a new piece for us, so I just went back to doing my workshop style which my kids always made growth. The last two years we did NWEA and I thought…I’m just going to keep doing that. They both indicated that their primary objection to the mandates was that they did not believe that the new system was better than their old practices. Ms. Lavin explained: So really what we are looking at is just our program for the most part and our program sucks. Like I said it’s very surface, it’s, it spirals so much that I don’t feel that you’re really delving, the kids are really getting into any one thing, well enough and we are just going on to the next thing. I feel like our anchor text are horrible. Like when we would do the units [in the past], we would actually go through a whole chapter book. The ability to be able to see characters through and change throughout a whole text and see them grow, is so huge. But [in Journeys], they have these short snippets of terrible like anchor text …they’re horrible, they’re not interesting, they’re not rich. So I’ve seen a huge change in my students’ interest in reading. Like I used to be able to get my students really into the Great Gilly Hopkins and learning about those characters or those really great novels like we’ll get into Roald Dahl. 169 By May, both teachers indicated that they were simply taking the learning targets from the Hillside curriculum and then teaching the concept using their instructional strategies from the former program, Reading Workshop. Ms. Lavin does so, because she believes that her practices deepen the concepts from the level described in the Journeys resource. These beliefs may have isolated Ms. Lavin and Ms. O’Neill from the rest of their schools’ literacy advice network. When examining Ms. Lavin and Ms. O’Neill’s position in their school’s advice network (as shown above in Figures 5.1 and 5.2), it is interesting to note that Ms. Lavin is on the outskirts of Willow’s literacy advice network. This position could indicate a lack of connection to the school’s overall literacy conversations. Ms. O’Neill is closer to the center of the network at Hickory Elementary School. This more central position could be the consequence of her close relationship with the literacy coach, who was a close friend. Both Ms. Lavin and Ms. O’Neill had over 10 years of teaching experience in the Hillside district. This tenure may explain some of their resistance and their feelings of being able to skirt away from district expectations. As Ms. O’Neill explained: I’m not breaking rules, but I am breaking rules. And in the situation I don’t care because there are too many other, I mean we have people retiring like nobody’s business right now and some I think that are not being as backed. So in the scheme of things, I think that I am a fairly good teacher and I decided that after a certain point that, no, I’m not worrying about it. I’m going to do what’s best for my kids and hopefully NWEA will play that out, but even then I don’t really care about that. I have kind of gone rogue. Well it’s the only way to keep myself sane! 170 For these experienced and knowledgeable teachers, they were not simply going to implement a system because district leadership told them to. They were not convinced by the experimental science assumption that the knowledge is in the program. Lakeside Implementation At Lakeside school district, teachers’ implementation of the systems and beliefs about the mandated literacy program seemed to be based more on which elementary school they were teaching in than their personal backgrounds and beliefs. In fact, teachers at both Cottonwood and Rose Elementary Schools shared that they appreciated the new literacy expectations entering their schools because they made the expectations black and white for what types of instructional strategies were considered high-quality. Ms. Parrish summarized this feeling in terms of the vision of high-quality instruction: I think I give you this answer every time, but you know there’s a very clear cut concise plan for teaching and it is so laid out for you on what the expectations are for you, in a classroom, as a teacher. That as long as you can read, I don’t think that it’s that difficult. This is what the teacher should be doing, this is what I should see the teacher doing, this is what I want to see the students doing. It doesn’t get any more clear that that. In interviews throughout the school year, Ms. Stewart and Ms. Johnson communicated similar beliefs regarding the clearer expectations around literacy instruction across the district. This common belief in the value of the mandated systems meant that teachers were less likely to openly resist the reform. However this belief did not ensure consistent implementation and maintenance of the new systems over the entire course of the school year. At Lakeside, teachers’ implementation and orientations toward the reform seemed to be based mostly on which school they taught in. Teachers could be classified into the following 171 categories: the Leaders and the Fading Teachers. The Leaders were those teachers at Cottonwood Elementary School, because as explained above, the teachers at Cottonwood introduced the CAFE and Daily 5 systems into the district. Consequently, overall, they implemented and maintained their use of these systems throughout the entire school year. The Fading Teachers refer to the teachers at Rose Elementary school. Overall, the teachers at Rose Elementary School implemented the systems at the beginning of the school year but their use faded by May and June. The Leaders – Cottonwood Elementary School – "This is working, let's keep doing it." (Ms. Johnson). The CAFE and Daily 5 literacy programs began in Cottonwood Elementary School. Ms. Parrish summarized the evolution of CAFE and Daily 5 in Cottonwood as: It started about five years ago, but it was kind of hit or miss in classrooms. Some people were using it, some people were not, now everybody has been instructed and you are held to it. Everybody has to use CAFE. You have to use teaching, you have to use the strategies to guide your teaching. It’s nice because in fourth and fifth grade, those strategies, I don’t have to figure out which strategy to figure out with, they are outlined within the Making Meaning program. As stated by Ms. Parrish in the quote above and described in more depth in the origin of the plan section, Cottonwood was the leading school in bringing the CAFE and Daily 5 systems to the district when Ms. Johnson and a kindergarten teacher attended a professional development and brought the practices back to the school. During the 2016-2017 school year, in all four observations that I completed with Ms. Johnson and Ms. Edmund, they taught using the CAFE and Daily 5 systems. Additionally, Ms. Parrish taught in alignment with the expectations that she and her teaching partner would be 172 teaching social studies or science using CAFE skills that district leaders agreed to (which was described above in the flexible monitoring section). Admittedly, Ms. Edmund’s implementation was at a more superficial level in that her individual conferencing did not seem to have the same intention and direction as Ms. Johnson’s, but overall she was using the systems. Figure 5.3 shows the informal literacy advice seeking network at Cottonwood Elementary School. Ms. Johnson and her Kindergarten colleague that spearheaded the introduction of CAFE and Daily 5 are labeled with asterisks. Interestingly, they were in two different subgroups; however, there existed many ties between groups. This placement of each expert teacher in a different subgroup may have helped spread the expertise in the literacy systems across the school and helped maintain the pressure to continue teaching to the system. Figure 5.3: Cottonwood Elementary Literacy Advice Network Figure 5.3 shows the informal ELA advice network at Cottonwood Elementary School. Interestingly, each teacher participating in the study was also from a different subgroup. Regardless of subgroup, they each maintained teaching to the mandated systems. For Ms. 173 Edmund, who may have been the most likely to abandon the practices, since she was not a leader of CAFE like Ms. Johnson or reading embedded in social studies and science, like Ms. Parrish, her proximity to the Kindergarten teacher who was also a leader in CAFE may have helped her maintain the systems and helped her find resources. Beyond the proximity of literacy experts to most teachers in the system, the principal at Cottonwood, Ms. Wilson also made efforts to put a focus on the expected literacy practices. As mentioned above, Ms. Wilson hosted a voluntary “reading summit” for teachers to learn more about the high-quality teaching practices. Ms. Parrish described the night as the best professional development that she received all year: I guess the best one would be our reading summit. I keep calling it, I don’t know really what to call it, the seminars that aren’t staffed, it was voluntary. It was after school and I think all but two staff members came. To me that says a lot…You didn’t have to go and we had almost everybody there. It really was a great way to grow this highly qualified instruction. We had to watch a video of a teacher that was teaching a high quality instruction lesson. Then we kind of had to fit and debrief and decide. There was a checklist of things that the teacher did and students did…The teacher did everything but one thing on that checklist. This reading summit helped the teachers to understand the process that the principal was going over in her HQI walkthroughs and understand the high-quality actions at a deeper level. After the summit, the principal also began giving feedback from HQI walkthroughs by leaving handwritten notes on that checklist. Notably, at Cottonwood the principal. Ms. Wilson, was also embedded in the subgroup with Ms. Johnson, which may both provide her with more access to 174 Ms. Johnson’s expertise as well as have allowed the teachers see Ms. Wilson more as an instructional leader than a monitor. Lastly during the 2016-2017 school year, Cottonwood Elementary School was named a Rewards School under the state’s accountability system due to the double-digit growth in test scores during the 2015-2016 school year. All three teachers attributed this success to the implementation of CAFE and Daily 5 at the school. Ms. Johnson shared this thought after hearing that they were recognized as a Rewards School: I really think that, first of all, our principal last year, we came to him with ideas. The kindergarten teacher went to this workshop and said, "We need to do this." He gave us full reign and said, "Go ahead." Never stopped us. He really empowered us and said, "If you think this is going to..." and you can see it started working and he just kept buying the supplies, he was really supportive of that. Once we started seeing results it fueled the fire to say, "This is working, let's keep doing it." That's what I think really was ... Now there's a lot of pressure though. Right now we're taking tests and we're all kind of freaking out thinking, "What the heck? Are we going to make this keep working? How's this going to keep going?" Right now there has been a little bit of pressure with regards to that. As Ms. Johnson shared, the success that they had with the program drove them to keep teaching to the program and to keep finding ways to learn and improve. Collectively, this recognition as a district leader in literacy practices, the accumulated experience with the practices and expertise, and the additional support provided by the principal seemed to create a school system in which teachers taught to the district expectations and did so in a way that recognized professional discretion and accepted variation. 175 The Fading Teachers – Rose Elementary School - “I feel like they can't leave it to us. Not that we can't handle it, but I think it's a lot to ask” (Ms. Olin). This was the first year that the CAFE and Daily 5 systems were mandated at Rose Elementary. While they had some experience experimenting with the systems during the 2015-2016 school year, this was their first year attempting to teach with fidelity. All three observed teachers began the school year teaching to the CAFE and Daily 5 systems. However, as the year progressed, the teachers at Rose had difficulties maintaining the instructional demands of teaching to the CAFE and Daily 5 systems. As will be explored further before, this could have been due to a lack of expertise and support as compared to Cottonwood Elementary School. It could also be related to the decline of a healthy professional community at Rose. In fact, by May, half of the teachers at Rose had made the decision to transfer out of the school for the 2017-2018 school year. First, Ms. Stewart described the changes that she made in her instructional practices at the beginning of the year in response to the new district expectations: I think we talked at our other meetings that this is the first year I've done the Daily Five and the CAFE with fidelity because it's the first year anyone has ever really sat me down and taught me how to do it. I think there's been more of an emphasis on the kids in small group, while I'm in small group, instead of like, before I may have just had a whole group lesson done, like I'd roll it out, "Okay, you work on this and now I'm going to pull these kids, and when you're done, you can read a book," or something like that. She and her colleagues at Rose did make an effort throughout the school year to teach CAFE and Daily 5 with fidelity; however they all had some concerns with the looseness of the systems. Both Ms. Stewart and Ms. Olin expressed some challenges with the loss of a basal and curriculum map to help them organize their units and lessons. Ms. Olin shared: 176 There's just no clear guideline of what exactly we're supposed to be doing. And I feel like, on one hand, it's professional liberty to teach what you think the kids need along with making sure it's aligned with the standards. But it would be nice to have a map to know what I should be doing when. Because sometimes I spend a really long time on something, because they like it and they're getting it and I'm going deeper, but then I'm like, "Oh my God, I have to get this done and I'm not gonna get it done now." You know what I mean?" Even with the “professional liberty” granted within the Lakeside system, they all taught using the CAFE and Daily 5 systems, with some minor modifications like Ms. Olin taught using three stations instead of five. This implementation level continued until May when they all fell back into their former practices. They taught to the system and altered their instructional practices like Ms. Stewart described above, because they all expressed a belief in the value of the district’s systems in promoting student growth in reading. In fact, as they got more experience with the systems, their belief in the systems grew. Ms. Clark shared that before this year, she had dabbled with the systems and then at the beginning of the 2016-2017 school year, she decided to really go for it and teach the first ten days in the structured way that the book recommended. Following that experience she noticed: I saw that the strategies were being used and of course as you grow in reading levels, accuracy's not so much as important at certain levels as comprehension is, or the expanded vocabulary for the higher-level kids. So, as we moved through growing in reading, we moved over to the newer strategies and then it came to a point where okay, we're going to stop and we're going to really dive into comprehension because one of the struggles that I've seen is they have comprehension, they can tell you but that deeper 177 level is not there. So they had a lot of modeling and a lot of practicing and now it's just kind of embedded in them so. But I put that all back on the way it was structured first thing when they first came in the door on the first day of school. Basically, because Ms. Stewart saw such success with the program after following it the first ten days, she kept these structures for the rest of the year. She shared that the program allowed her to really “focus the lesson, whatever that group needed, the lesson would be focused around them.” This belief in the mandated program was not found in any implementation group at Hillside. Even with this belief, by May all three teachers at Rose were teaching in a more whole- class, direct instruction format. All three teachers shared that they were having troubles with student engagement and behavior. Because of these challenges and possibly some fatigue at the end of the school year, they were falling into their old practices and simply trying to keep students engaged in the lesson. Ms. Stewart explained that she taught a whole class lesson in which students write letters to their future fourth grade teachers, because “it's just one of the things I do at the end of the year. As you dwindle down, you start looking for things that are still academic, yet a little bit fun.” Ms. Olin described a hope that next year, district leaders should be: giving us some kind of map, or guideline, of when to do what. That is something I hope they're going to do. I feel like they can't leave it to us. Not that we can't handle it, but I think it's a lot to ask. This stress of forming their own instructional program at Rose may had been heightened because they did not have access to those Cottonwood teachers, like Ms. Johnson, who were invested, passionate, and highly knowledge about the programs and other literacy resources. 178 Figure 5.4: Rose Elementary Literacy Advice Network An additional challenge in maintaining the literacy systems may have been in the lack of supports that the teachers felt that they received to support their literacy instruction. As shown in Figure 5.4, Ms. Clark and Ms. Olin were both located in a subgroup of fourth and fifth grade teachers who seem to be rather isolated from their colleagues. Additionally, the principal, Ms. Reed, was isolated from her teachers making it difficult for her to serve as an instructional leader. Lastly, both the principal and Ms. Stewart reflected that by the end of the year the professional culture had deteriorated at Rose. Ms. Stewart shared: There’s definitely been some challenges among personalities this year…it’s more clicky. It’s like, I’ll admit it. I have my people. I’m not, I’m still friendly to hose people that I wouldn’t consider my people. Ms. Reed in her spring interview also shared a regret in that she did not take enough time to develop relationships with teachers this year. As stated above, by May, half of the teachers at Rose had made the decision to transfer out of the school for the 2017-2018 school year. This lack 179 of commitment to the school may have also contributed to the setback in teaching to the new literacy systems. Role of Knowledge in Implementing Tendencies When analyzing the data, the theme of background knowledge, particularly reading pedagogical knowledge emerged as a common rationale for either resisting, adapting, or implementing the mandated reform. Most interestingly is that, between the two districts this background knowledge seemed to lead teachers to make different decisions regarding implementation. Hillside School District Faithful Implementers- high implementers Adjusted Implementers – middle implementers Resisters - low implementers No Specialized Reading Training Ms. Miller Ms. Morey Ms. DeSutter Ms. O’Neill Lakeside School District Leaders- high implementers Ms. Edmund Ms. Parrish Undergraduate Minor or Major in Reading Advanced Literacy Training Ms. Trainor (working on National Boards in Literacy) Ms. Lavin (extensive training in workshop model through ISD) Ms. Johnson (CAFE trained and Reading with the Experts and Masters in reading) Fading Implementers- middle implementers Ms. Olin Ms. Clark Ms. Stewart Table 5.2: Advanced Literacy Instructional Knowledge and Teacher Implementation Practices 180 Advanced reading pedagogical content knowledge seemed to be related to a teachers’ orientation to their district’s literacy program. As shown in Table 5.2, in Hillside, more knowledge about literacy resulted in more subversive behaviors. During the 2016-2017 school year, Ms. Trainor was going up for her National Board Certification in Literacy. Applying for National Board Certification is an extensive process requiring a deep pedagogical knowledge in best practices for reading instruction and differentiation. This knowledge made it difficult for her simply to teach to Journeys and the Hillside Curriculum. She shared, "I use Journeys as a resource, but my background is so strong in that, that everything ends up coming out in the structure of reading workshop, even if I don't want it to." Likewise, Ms. Lavin shared that her deep knowledge in best practices in literacy and her advanced training in the Reading Workshop model made it difficult for her to teach to the basal series. Beyond experience, Ms. Lavin was also a leader of the Reading Workshop system that the district had formerly mandated. She spent her first seven years attending additional training on the workshop model and helped to develop the units. This extensive training and involvement in the previous system really influenced her perceptions of the new system as being worse than the previous one. On the other hand, in Lakeside, stronger knowledge in best practices led to a deeper implementation of the district’s systems. Ms. Johnson not only had attended the training on the CAFE model of teaching, she also held a Master’s degree in reading instruction and continued to attend professional development called “Reading with the Experts” provided through her ISD. This extensive knowledge allowed her to not only implement the systems, but to teach them with fidelity in a manner that adjusted to the needs of individual students and small groups of students. She attributed this continued learning to her ability to better teach the systems: 181 I did it last year, too, but I’m even doing it more now. I’m not, it’s more individualized and the more I learn about how to teach the strategies more deeply, like with CAFE they show you how to do it but the Serravalo book will take it like I’m teaching them how to summarize. Well, once you get to summarizing there’s different levels of summarizing and how to take it to a different level. And the Jennifer Serravalo book really helps to do, to find quick ways to make them read a little bit more for better understanding. In other words, Ms. Johnson’s deeper level of knowledge helped her to be able to teach the strategies at a deeper and more individualized manner. In her observations, that was also evident in how she differentiated her individual conferencing conversations and her small group skill lessons to teach the students reading skills at their levels. These findings could indicate that more scripted programs may be better implemented by teachers with less experience and specialized knowledge; while more professional-based programs relying on systems and professional discretion may need a more experienced and knowledgeable staff in order to implement and sustain. Quantitative Examination of Implementation Examining teacher survey results on reading instructional practices, planning practices, collaborative practices, and feelings of fit in their school, I examine what related to a teacher teaching reading in a manner aligned with the district’s literacy program as defined by their instructional framework. 182 Hickory Willow Cottonwood Rose 12 3.1208 3.125 0.5353 1.6 2.4 4 N Mean Median Std. Deviation Range Minimum Maximum 9 2.6928 3.235 1.6869 4.647 0 4.647 9 3.752 3.647 0.6327 1.7647 2.8823 4.6471 10 3.4006 3.3 0.6936 2.4 2.6 5 Table 5.3:Spring Aligned Teaching Descriptive Statistics Table 5.3 shows descriptive statistics for each school around the aligned teaching of their teachers in the spring, as measured by items measuring literacy practices found on each district’s implementation guide ranging from 5 for all 5 previous lessons to 0 not taught in the last 5 lessons. Table 5.1 shows the descriptive statistics for the outcome variable of teaching reading in a manner aligned with the district’s implementation guide in the spring. Overall, Hickory had the lowest average number of lessons with teaching aligned to the district’s guidelines. Willow has the highest average. None of these means are different from the others at a statistically significant level, most likely due to the small sample size. 183 Model 1 Fall Aligned Teaching Exposure Term Exposure Term- Missing Flag Planning using district resources Beliefs that the Policies are Coherent Feelings of Fit in School Hillside District Early Elementary Teacher (Grades K-2) Yrs Experience in Education Reading Background Gender Constant R2 N -0.16 (0.35) 0.63* (0.34) -0.01 (0.02) 0.29 (0.34) 0.71 (0.56) 2.39 (0.62) 0.20 40 Model 5 0.59*** (0.14) -0.02 (0.05) -0.23 (0.54) 0.34** (0.16) -0.16 (0.25) 0.16 (0.21) 0.25 (0.26) 0.63** (0.26) 0.0006 (0.01) -0.06 (0.26) 0.46 (0.40) -0.64403 (0.8578) 0.66 38 Model 2 0.76*** (0.11) Model 3 0.75*** (0.11) -0.01 (0.05) -0.09 (0.53) Model 4 0.59*** (0.13) -0.01 (0.04) -0.09 (0.52) 0.30* (0.15) 0.01 (0.22) 0.65*** (0.22) 0.001 (0.01) 0.04 (0.84) 0.45 (0.23) -0.054 (0.53) 0.68 40 0.002 (0.24) 0.67** (0.25) 0.001 (0.01) 0.03 (0.24) 0.45 (0.39) 0.02 (0.72) 0.68 40 0.19 (0.24) 0.63** (0.25) 0.0003 (0.01) 0.04 (0.23) 0.45 (0.38) -0.41 (0.77) 0.64 39 ***p<0.01; **p<.0.05; *p<0.10 Table 5.4: Summary of Linear Regression Results for Spring Aligned Reading Instruction Table 5.4 shows results from a linear regression analysis examining factors related to teaching reading in a manner aligned with district expectations in the spring. Model 1 represented a base model with only teacher characteristics. In this model, only the grade a teacher taught marginally statistically significant with teaching in grades K, 1, and 2 being positively related with implementing spring aligned teaching practices. Model 2 added in a 184 measure of a teachers’ previous aligned reading teaching practices in the fall of 2016. In this model, a teacher’s previous practices were statistically significant and positively related to spring instructional practices as were the grade level she teaches. Again, teaching in the early grades remained to be positively related to teaching in an aligned manner in the spring. Model 3 added a consideration of a teacher’s informal network. The network exposure term was not statistically significant; however, a teacher’s fall teaching practices and teaching in an early grade both maintained their positive relationship to spring aligned teaching. Model 4 added the measure of a teacher’s planning practices. A teacher’s use of the district provided resources to plan her reading lessons was positively related to her teaching in an aligned manner in the Spring. In Lakeside, this construct is measured as using the CAFE book and the Making Meaning book. In Hillside, this construct is measured as using the Journeys textbook and the curriculum. Lastly, Model 4 added teacher’s belief regarding the implementing policies and feelings of belonging in her school. Neither of these constructs were statistically significant. Overall, it appears that the factors most related to whether a teacher teaches reading in an aligned manner in the spring was her former teaching practices, her planning practices, and the grades she taught. These models could indicate that it may be easier for teachers teaching in the lower grades to change their reading instructional practices and sustain them, more so than a teacher in the upper elementary grades. This could indicate that altering reading instructional practices for content addressed in kindergarten, first, and second grade was somehow easier than the more complex texts and concepts covered in upper elementary school. Summary This chapter addresses research questions three and four: (3) How is the theory of change enacted in each district? How does this enactment reflect administrators’ orientations toward 185 change? and (4) To what degree and in what ways does the enactment of the theory of change influence teacher instructional practices and planning in each school? Administrative beliefs regarding change influenced both the literacy program chosen to implement district-wide as well as the implementation process. In Hillside, the administrators with more experimental science assumptions chose to mandate a more scripted literacy program that asked teachers to put their faith in the program. The administrators then installed numerous monitoring systems that ensured teachers faithful implementation of the program, such as observations from their principals and coaches as well as data collection expectations for each reading unit. Additionally, they limited access to data and outside resources in order to reduce variation. On the other hand, at Lakeside, the administrators’ improvement science assumptions led to the decision to adopt with additional training, the practices that teachers were having success with at Cottonwood Elementary School. This program was more of a system of how to teach reading and relied on teachers making informed decisions to target reading skills instruction to meet the needs of individual students and small groups. The implementation systems that they installed allowed for and supported variation between classrooms and included teachers in plan-do-study-act cycles when analyzing instructional and student achievement data. These program and implementation process decisions seemed to also influence teachers’ beliefs about the reforms and implementation practices. Overall, implementation patterns seemed to differ by district. In Hillside, the participating teachers shared different beliefs regarding the reforms and implementation patterns based on their individual background, such as experience and pedagogical content knowledge in literacy. At Lakeside, implementation seemed to be influenced more by school-based influences. At Cottonwood, teachers worked collectively with their principal and teacher leaders to build on 186 their previous experiences with the CAFE and Daily 5 programs. This collective work resulted in all three teachers teaching to the systems into May. At Rose, they did not enjoy the same support from their principal and there were no CAFE teacher experts working in the school. Interestingly a teacher’s individual knowledge and experience level seemed to influence their beliefs regarding the reforms in opposite ways based on which district they were teaching in. Teachers at the Hillside district with more experience and literacy pedagogical knowledge expressed more intention in modifying the expectations to fit best practices, as Ms. Trainor worked to do. Alternatively, the more experienced, Ms. O’Neill, and more knowledgeable, Ms. Lavin, also outright resisted the reform and chose to teach the learning targets using their prior practices. At Lakeside, Ms. Johnson represented the ability of a knowledgeable and experienced teacher to utilize their professional knowledge using the systems in order to differentiate and meet the needs of her individual students. Quantitative analyses of factors relating to spring reading teaching practices aligned with district expectations showed positive relationships with a teacher’s fall teaching practices, her planning practices with district-provided resources, and teaching at the K-2 level. These results may indicate that in terms of reading instruction, it may be easier for K-2 teachers to change their instructional strategies than teachers teaching in upper elementary where students engage with more complex texts and literacy concepts. 187 CHAPTER 6: DISCUSSION, IMPLICATIONS, AND CONCLUSION Summary of Findings This dissertation analyzed the implementation of the Blueprint for Systemic Reconfiguration at the district-level, school-level, and classroom-level as leaders attempted to influence teachers’ instructional practices through systems reform. The purpose of this dissertation was to analyze the theory of action of the Blueprint for Systemic Reconfiguration and its connections to improvement science, explore how district and school leaders took up and implemented the Blueprint into their school organizations, and lastly to examine how systems reform in this case influenced teacher reading instructional practices at scale. Blueprint for Systemic Reconfiguration Theory of Action The Blueprint for Systemic Reconfiguration draws on improvement science paradigms of change to create a framework for school turnaround. Overall, the Blueprint focused on systems reform in order to improve teacher and student performance. District leaders, meeting weekly as the district network (DN), design improvements and clarified language around district systems addressing human resources, curriculum and instruction, and supporting students’ non-academic needs. This focus on systems was congruent with the improvement science paradigm in that organizational results are the consequence of how the systems of that organization function (Perla et al., 2013). Through driver systems like the Performance Management Driver (PMT) and the Problem-Solving Driver (PSP), district and school leaders had the opportunity to engage in a form of plan-do-study-act cycles at least monthly to assess the impact of district policies on student achievement (running records, NWEA) and behavior (attendance, office referrals) data. 188 Leadership Assumptions of Change and Implementation Superintendents in two districts worked to implement Blueprint for Systemic Reconfiguration into their school organizations. Both leaders followed the guidance from MI Excel to implement components and systems into their organization according to a provided timeline. However, each superintendent held contrasting assumptions about change. The superintendent at Hillside, Ms. King, primarily held experimental science assumptions about change in which she valued fidelity to programs and reducing variation in instruction between classrooms as much as possible. On the other hand, the superintendent at Lakeside, Mr. Dixon, primarily held more improvement science assumptions of change in which he valued honoring teacher professional knowledge. These assumptions resulted in an acceptance of variation in practice across classrooms. These differing assumptions of change resulted in different choices when implementing Blueprint systems into their districts. At Hillside, district leaders decided to mandate a basal series with a district-made curriculum; they restricted access to the data to Building Network (BN) members for Problem-Solving Protocol (PSP) analyses; principals completed High-Quality Instruction (HQI) walkthroughs but did not leave feedback until April; and lastly two district reading coaches provided teachers support, but also reported back to central office on teacher implementation practices. At Lakeside, district leaders mandated literacy systems of instruction, CAFE and Daily 5; PSP data analyses were completed at full staff meetings allowing all teachers to view and analyze their data; and principals completed HQI walkthroughs and regularly left some sort of feedback. 189 Influencing Teaching Practices at Scale In general, the Blueprint for Systemic Reconfiguration attempted to improve instructional practices at scale throughout a school organization. As Chandler and Frank (2015) describe, “While what truly matters to improving student performance occurs in the classroom between the teacher, student, and content, comprehensive school reform efforts have failed in large part because they have focused on improving student learning and teacher performance one classroom at a time.” Through the decisions made by district leaders when addressing the “Instructional Infrastructure” component of the Blueprint, teachers were asked to teach reading in a different manner than in prevous years. At Hillside, district leaders asked teachers to teach to a basal series and district-created curriculum and pacing guide. At Lakeside, district leaders asked teachers to teach using new techniques from the CAFE and Daily 5 literacy systems. For teachers in both districts, the 2016-2017 school year was their first year attempting to teach to these new expectations. In both districts, leaders also implemented supports to support teachers attempting these new teaching strategies. In both districts, principals completed HQI walkthroughs to monitor and, at Lakeside, to provide teachers with feedback on their instructional practices. At Hillside, district leaders also hired coaches that pushed into classroom and assisted with reading teaching practices like guided reading groups and shared reading. These coaches also reported back to central office leaders about teacher progress with the curriculum and the expected teacher strategies. At Lakeside, teachers were able to analyze their student achievement data in reading in the form of running records and NWEA scores at PSP meetings. These meetings provided teachers with an opportunity to design interventions for students who were below grade level or had plateaued in their reading. 190 Interestingly, there seemed to be differences in how teachers with advanced expertise in reading instruction reacted to district expectations. At Hillside, teachers with more teaching experience and expertise in reading instruction seemed more likely to resist components of the district’s expectations for literacy instruction. At Hillside, the teachers in their first year of teaching, were the teachers that implemented the district’s expectations with fidelity. I labeled these teachers as the “faithful implementers” as they simply attempted to survive their first year of teaching by abiding by the district’s rules and teaching to the expectations. Hillside teachers working collaboratively in a 2nd grade team at Willow Elementary School “adaptively implemented” the district curriculum and basal series content. Ms. Trainor who planned the reading lessons for the team was also working on her Master’s in reading instruction and towards her National Board Certification in Literacy. Drawing on this advanced knowledge, she would rework district expectations into the workshop model of instruction that she knew from experience and coursework were the most effective in teaching reading. Lastly, two experienced teachers were “resisters” of the district’s expectations. Ms. Lavin had been heavily trained in the previous system of Reading Workshop and felt like the new district expectations were mandating less effective methods. Consequently, by the spring, she reverted back to her previous instructional strategies. At Lakeside, advanced expertise in Reading instruction seemed related to more advanced implementation of the district’s expectations. All six focus teachers expressed a belief in the value of the district systems and attempted to implement CAFE and Daily 5 into their instructional strategies. Teachers at Rose Elementary attempted to implement the systems throughout the year. However, as the year progressed, the demand of searching for materials caused them all to revert to whole-class instructional techniques using Scholastic magazines or 191 lesson plans from previous years. At Cottonwood, teachers received advanced training in CAFE through support from the original teachers that brought the methods to the district and the principal’s incorporation of reading study into staff meetings. This resulted in teachers implementing the practices at a deeper level and sustaining these changes throughout the entire school year. School Turnaround and Improvement Science This dissertation provides another example of improvement science principles attempted in schools. The theory of action as well as the assumptions of change embedded in the Blueprint for Systemic Reconfiguration provide a new option for school turnaround efforts that rely on social capital and improvement science assumptions of change. For the past two decades, policymakers and researchers have worked to improve student performance in low-performing schools (Duke, 2012). Under NCLB, schools failing to make adequate yearly progress were given the options to either: (1) reopen as a charter school, (2) reopen under the operations of an outside organization, (3) reopen under the operations of the state, (4) replace the principal and a large proportion of staff, or (5) undergo other major restructuring with improved professional development, curricula, or other related interventions (Duke, 2012). Similar policies remained under the Obama administration where schools labeled as “persistently low-achieving” (schools ranked in the bottom 5% of a state’s school in student achievement) were given four options for “turning around” student achievement. Overall, these school turnaround policies under NCLB and RTTT prioritized theories of action for improving schools rooted in the assumptions that poor student achievement was a human capital problem in which the solution relied on new leadership and/or better professional development to improve teachers (Duke, 2012; Rogers-Chapman, 2013). This focus on human 192 capital solutions often led to demoralized teachers, which led to some contrary results as the newly hired teachers tended to have less experience and oftentimes were alternatively certified (Hamilton et al., 2014; Rice & Malen, 2003). Moreover, the new curriculum and other instructional reforms implemented under turnaround regularly resulted in professional environments where teachers felt monitored for compliance to arbitrary rules and overly scripted curriculum (Cucchiari et al., 2015). In other words, the actual instructional policies implemented under the turnaround reforms followed the experimental paradigms for change. In the last decade, researchers and some policymakers have worked to incorporate improvement science principles into school improvement efforts (Bryk et al., 2011; Lewis, 2015). Drawing from social capital assumptions of change, rather than human capital, improvement science prioritizes system reform where local stakeholders are able to address organizational conditions through plan-do-study-act (PDSA) cycles (Langley et al., 2009). These cycles rely on networks of people coming together to design an improvement, testing the improvement in the local environment, analyzing data on the test, and lastly deciding whether to implement at scale the change, modify the change and retest, or abandon the change idea (Bryk et al., 2011; Langley et al., 2009). In school settings, Bryk and colleagues (2011) have used networked improvement communities (NICs) to facilitate the completion of these PDSA cycles for designing and testing improvements in the local context. Embedded in this paradigm of change is the assumption that local actors have the knowledge and potential to integrate their knowledge of practice and context into the creation and implementation of potential improvements to organizational operations (Bryk et al., 2015). The Blueprint for Systemic Reconfiguration drew on improvement science principles to offer an alternative to these traditional turnaround policies. The Blueprint allowed school and 193 district leaders to design systems and implement instructional programs that respond to local contexts. The Blueprint provided an alternative organization of improvement science assumptions into school improvement to compliment the work being done in NICs and lesson study. The first finding of this dissertation explored how the improvement science assumptions can be organized and enacted through the Blueprint’s theory of action. District and school leaders used the Blueprint framework and tools to address the systems within their districts and to influence teachers’ reading instructional practices. In the Blueprint, the work of the district and building networks mirrored the literature on networked improvement communities (Bryk et al., 2015). Like NICs, the Blueprint’s networks engage a diverse group of stakeholders from around the district and school building to exchange their knowledge and perspectives. The networks worked to develop new organizational protocols and practices aligned with the pre-established Blueprint system, try them out in the organization, and then come together to adjust them to fit to their organizational needs. Through the use of networks employing practices similar to those in the networked improvement communities literature, districts strived to collaboratively construct a district vision of high-quality instruction, implement it throughout all district classrooms, and respond to obstacles in implementation at both building network and district network meetings (Frank & Chandler 2016). Interestingly, the second finding of this dissertation pointed to the ability of school leaders to modify the improvement science framework, as represented by the Blueprint, to their assumptions about enacting change. In Hillside, school leaders utilized the Blueprint to implement changes into the district modeled under experimental science assumptions of change. This resulted in teachers feeling stressed and monitored like those teachers working in schools attempting the traditional school turnaround programs (Rice & Malen, 2003; Cucchiari et al., 194 2015). On the other hand, in Lakeside like with the BTENs (Hannan et al., 2015), when leaders implemented the Blueprint following improvement science assumptions the fear and stress that teachers often feel when leaders introduce a new initiative was ameliorated. District and school leaders drew on improvement science principles to promote buy-in with the district’s reading policies, to involve all stakeholders in PDSA cycles in PSP meetings, and to provide feedback to teachers regularly and consistently on their reading instructional practices. This finding suggests that district and school leaders may need some training on the assumptions and processes in improvement science in order to lead such an effort. While MI Excel, the organizers of the Blueprint, did offer many trainings to district and school leaders, these professional learning opportunities centered primarily on the systems and drivers in the Blueprint, what they were and how they were designed to work, rather than a focus on the assumptions of how change would occur throughout the entire process and the perspectives that leaders should take while facilitating this change. This lack of learning about improvement science assumptions resulted in differing implementation choices between districts. For example, Hillside leaders mandated a reading curriculum to be taught with fidelity across the district. This communicated a belief of the knowledge for implementing residing in the program. While at Lakeside, leaders mandated a system for reading instruction that provided for variation across the district and modifications in terms of the resources incorporated into the system by each teacher. Similar to the findings of Hannan and colleagues (2015) from the BTEN network’s success in implementing change in principal feedback cycles in two elementary schools, this dissertation points to the potential of improvement science principles to affect change in school routines. At Lakeside, district and school leaders observed similar success in influencing teacher reading instructional practices. All six observed teachers, implemented the CAFE and Daily 5 195 routines into their classrooms with some modifications to meet the developmental, ability levels, and social-behavioral needs of their students. This study provides further evidence of the potential of improvement science principles to facilitate change in schools. A problem exposed by this research is between the idea of rapid school turnaround and improvement science, which as Bryk and colleagues (2015) describe as the process to “learn fast to implement well.” The Blueprint called for district and school leaders to implement multiple programs at an accelerated timeline to improve systems in a district with the hope of improving student and teacher performance. However, this implementation of multiple programs at once made attributing causal statements to changes in data difficult, if not impossible. Improvement science calls for a more focused and intentional process to assess each program through PDSA cycles. This implies that the Blueprint for Systemic Reconfiguration could be improved by encouraging districts to focus on addressing one system at a time and allowing for time to test interventions for impact in each district’s context. This would place less of a focus on rapid implementation and more on the necessity to learn about programs and context in order to implement systems that can have long-standing success. What does good teaching at scale mean? Most research addressing the problem of scale focuses on theory, leadership practices, and instructional systems. Work by Elmore (1996) and Cobb and Jackson (2011) has outlined theories and frameworks around how to organize the process of impacting instruction at scale throughout school organizations. Meanwhile, improvement science researchers such as Bryk and colleagues (2015), Langley and others (2009), and Hannan et al. (2015) have argued for the value of the improvement science paradigm in education and the promise of the NIC structure to design and evaluate change efforts. Others have examined implementing programs at scale by 196 solely using quantitative measures which while useful for measuring impact cannot examine teacher thinking around implementation choices (Stein, Berends, Fuchs, McMaster, & Sáenz, 2008). This dissertation provides some nuance to the literature by qualitatively considering teachers’ responses to efforts to influence practice at scale. Depending on a teacher’s previous experiences and expertise in the area, their response to the effort may differ. By examining the scaling up effort at the teacher level, it is clear that there is more to the work than simply organizing the theory of action using the best framework or administrative technique. The improvement science paradigm makes clear that in order to create change, leaders must build consensus around the importance of making a change and in the choice of change to be made (Lewis, 2015). At Lakeside, all six teachers shared a belief in the value of the systems that were being put in place. Leaders built this consensus by choosing to implement at scale a system that teachers at Cottonwood believed in from their advanced training and had already found some success with in terms of improving student achievement. Furthermore, they offered training at the beginning of the year from a combination of CAFE program officials and teachers who had already implemented the program. Lastly, they provided time during the second half of the previous school year for teachers to experiment with the CAFE and Daily 5 systems before they were to be held accountable to them. By contrast, for the more experienced and knowledgeable teachers at Hillside, teaching to a basal series and the district curriculum did not feel like an improvement. The two “resister” teachers outright disagreed with the district’s expectations. Even though teachers had the opportunity to serve on the curriculum writing committee, they still did not believe in the program’s value. For example, Ms. O’Neill helped write the curriculum and still disagreed with its value. For both Ms. Lavin and Ms. O’Neill scaling up a basal series was an inferior practice to 197 what they were doing before. For these teachers, this belief caused them to outright resist implementing the program and instead to teach the mandated objectives using their prior instructional strategies. For Ms. Trainor, her advanced literacy training also made it difficult for her to teach using the more direct instruction techniques she felt that the basal series and curriculum required. Like Ms. Lavin and Ms. O’Neill, she found alternative ways to use the district curriculum and basal series in an adapted workshop-like manner. Overall, these findings reinforce the improvement science assumption that in order to scale up instructional programs that leaders must work to build consensus among teachers, allow for experimentation by teachers and variation across classrooms, and respect the professional knowledge and judgement that teachers bring to their work (Lewis, 2015). Furthermore, incorporating improvement science assumptions into school improvement efforts raises a question of what influencing teaching at scale means under this paradigm. Traditionally implementation at scale has referred to implementing an innovation in more than a small set of classrooms and sustaining this change (Elmore, 1996). Over the past few decades under policy environments created by NCLB and RTTT, the concept of implementing at scale has come to signify implementing “best practices” at scale, which has resulted in hard-working and well-meaning teachers implementing programs that focus on uniformity. Unfortunately, the programs that are easiest to scale uniformly are those “that require the least depth of preparation and practice” (Elmore, 2016, p. 531). At Hillside, their leaders’ attempt to improve instruction at scale embodied these original assumptions about bringing instructional practices to scale. District leaders chose to implement a more traditional program, Journeys, along with a district-provided curriculum that provided pacing information and essential questions to follow. This idea of scaling up “evidence-based” 198 programs was consistent with experimental science assumptions in which teachers are required to teach the programs with fidelity (Lewis, 2015). At Hillside, leaders emphasized teaching to the mandated program without modifications. Both Ms. Trainor and Ms. Miller expressed uncertainty about their ability to incorporate outside resources. Unfortunately this notion of implementing “evidence-based” programs takes out the inquiry component of the improvement science paradigm because the inquiry has been completed before the teachers had access to the program (Anderson & Herr, 2011). As research and the experience of teachers at Hillside shows, teaching to a previously designated “evidence-based” program limited the opportunities for teachers to incorporate their knowledge of local contexts into their instruction. This study shows that when leaders mandate this type of instructional practice with experienced and knowledgeable teachers, they can push back on these expectations in intentional ways in order to maintain what they consider to be good teaching. In contrast, leaders at Lakeside attempted to bring to scale a program that relied on teachers being able to incorporate their knowledge of their local context and students’ abilities and interests with the literacy systems of CAFE and Daily 5. This emphasis on the importance of the teachers’ knowledge and the embrace of variation makes this program more consistent with the improvement science paradigm (Lewis, 2015). In a recent essay, Elmore (2016) argues that implementation at scale should focus on understanding which ideas can work in diverse environments. He argues that this focus will require a shift to the following characteristics: (1) basing learning on designs reflecting the theory and science of how humans learn, (2) focusing on the processes required to adjust powerful ideas to diverse contexts, instead of focusing on ‘universal’ prescriptions, (3) considering the multiple and promising adaptions to diverse populations, instead of universal prescriptions for learning, (4) sharing of expert knowledge in 199 simple manners in ways accessible to both students and teachers, and (5) expectations of and openness to surprising effects and patterns from divergent thinkers (p. 533-534). The present study provides evidence that the notion of scaling up under the improvement science paradigm must be thought of differently from implementing a program. Leaders at Lakeside chose to mandate a program that allowed for teachers to use their knowledge and experience to modify the program and for variation between classrooms in practices. Hannan and colleagues (2015) describe this new notion of scale up under improvement science as “adaptive integration.” They define adaptive integration as “using improvement research to integrate a standard work process into new contexts, which is critical for learning how to scale improvements.” (p. 496). In the case at Lakeside, adaptive integration involved learning how the standard work process of CAFE and Daily 5 could work across district classrooms. Consistent with improvement science assumptions this included allowing for variation in practice. Both district-level and school-level administration supported the idea of upper elementary teachers at Cottonwood to adapt the expectations to teach the literacy systems using social studies and science materials. At times this meant straying from the Daily 5 expectations, so that students could more deeply engage in the social studies content. For example, in Ms. Parrish’s class, I observed a mini-lesson in which she taught the CAFE comprehension strategy of “finding the main idea and supporting details” and then students worked in small groups with that strategy to make sense of a leveled reader on explorers. Throughout the school year, Ms Parrish and her colleague collected and analyzed student data on both reading and science/social studies scores through Running Records and NWEA assessments, to analyze the effect of their modification and make adjustments. This dissertation adds to the research of Hannan and colleagues (2015) about what adaptive integration of programs into schools and districts can look like. 200 Overall, as this dissertation shows the notion of scaling up under improvement science assumptions must intentionally involve teachers in a more active role. Teachers will need to think critically about programs and instructional systems. They will need to make modifications to the existing and new systems to meet the needs of their local context and the students in their classroom. Throughout their instruction, teachers must collect data on the effectiveness of the instructional programs and their modifications on student learning. They then must analyze the data and decide on any changes they should make to their current practices or how to continue with their current modifications. As we see in Lakeside, these modifications must also be shared with colleagues to influence the work of other teachers in the school and around the district. Implications Overall implications from this study suggest that improvement science assumptions may provide a promising new way to consider the work of school turnaround and influencing instruction at scale. Additional implications suggest that the orientations of leadership toward the change process may be pivotal to the implementation and effect of change efforts. This study has implications for the work of policymakers, school practitioners, and education researchers. Implications for Policymakers A persistent problem in education is in the inability to improve underperforming schools. The improvement science paradigm provides a departure from the experimental science paradigm that has been attempted for the past 20 years. This shift will require policymakers to consider not just “What works?” but “What works? For whom? Under what circumstances?” (Bryk et al., 2015; Slavin, 2002). This dissertation offers a first examination of a school turnaround framework built primarily under improvement science assumptions. The Blueprint for Systemic Reconfiguration offers a template for how school turnaround efforts could be 201 redesigned to reflect the improvement science methods that Bryk and colleagues (2015) have found successful in impacting remedial mathematics completion rates at community colleges and Hannan and colleagues (2015) found effective in improving principal feedback cycles. While it is too soon to measure the impact on student learning, since research argues that reforms should be given at least three years before assessing impact (Frank, 2014). This study shows the promise of the reform in impacting teacher instructional practices. At Lakeside in particular, teachers engaged in CAFE and Daily 5 literacy systems in ways that customized the programs to meet the needs of the students in their classrooms. More generally, this study indicates that policymakers should be more thoughtful about local contexts when designing policy. Policymakers should recognize that lockstep programs and prescriptions may not have the desired effects in every district, school, and classroom. The administration and teachers at Lakeside show us that when mandates follow improvement science assumptions and allow for variation, implementers are more likely to take up the initiative and implement it in a way that reflects their expertise and context. The Blueprint for Systemic Reconfiguration provides a model in that it provides a framework for change that district and school leaders can use to design interventions that match their needs and contexts. Along with the framework, leaders of the Blueprint provide support and advice for leaders as they work to design and implement their interventions. This combination of a framework allowing for flexibility and expert support could provide an example for how state-level policy can be responsive to local conditions. Implications for School Practitioners School Turnaround. School turnaround under improvement science assumptions will require practitioners to think differently about their roles in leading this change. Under 202 improvement science-based school turnaround school leaders will no longer be compliance monitors of decided programs, but instead will be facilitators of critical conversations in NIC- like environments and will be data analysts for the frequent rounds of data being collected as a regular part of school operations. Teachers will also have a more active role in the process. Teachers will no longer be passive implementers of programs decided higher up in the bureaucracy. Instead, school turnaround in the improvement science paradigm will require teachers to think critically about how to modify programs to meet the needs of students in their classrooms. They will have to collect meaningful data to engage in PDSA cycles in order to analyze the effectiveness of their modifications, and they will have to collaborate with colleagues to both analyze their data and share their modification strategies with others. These new roles will not come naturally. As shown by Hillside and Lakeside leaders and teachers even when implementing a system like the Blueprint for Systemic Reconfiguration, built on improvement science assumptions, it is possible to fall into the experimental science paradigm of faithfully implementing programs and monitoring implementation. Both teachers and leaders will need opportunities for professional learning around improvement science assumptions and methods. In Bryk and colleagues’ (2015) NIC organization this training occurs through the NIC facilitator; however other methods may be possible. The Blueprint offers districts consultants through their local intermediate school district. This role may need to be expanded to include providing improvement science training and facilitating collegial interactions more consistent with improvement science assumptions. Scaling Up Instructional Strategies. As described above, scaling up instructional strategies under the improvement science paradigm will also require new thinking and actions 203 from school leaders and teachers. This notion of scaling up will require school and district leaders to reconsider how they evaluate a teacher’s compliance with district mandates. In the old models of scaling up, teachers would follow a scripted curriculum and would be expected to be on the same lesson at the same time and on the same day (Elmore, 2016). At Hillside, school leaders attempted to scale up instructional strategies under these assumptions through mandating and monitoring the implementation of a basal series, district curriculum, and master schedule. On the other hand, in an improvement science system, like at Lakeside, scaling up will look differently in every classroom in order to respond to the needs of individual students in the classroom. Teachers may be teaching different strategies, but if they are teaching literacy using strategies consistent with the vision of high-quality reading instruction and the accompanying systems like CAFE then they are following district expectations. From a practitioner’s point of view, this different paradigm of scaling up may require different reflections of what aspects of instruction should be mandated and what should be left to professional discretion. Mr. Dixon, the Lakeside superintendent, described how he viewed his role as a district leader as deciding the aspects of his work in which he was an oak tree and in which he was a willow tree: And you know I respect the heck out of them, but there’s just sometimes I say all the time it’s the willow and the oak. And I say that all the time. There are many decisions in life that are the willow and they’re discretionary and you can bend like a willow tree and then there’s those very few times in life where it’s a core value and you just cannot bend. At Lakeside, district and school leaders acted as an oak in reference to the instructional strategies of CAFE, the Daily 5 structure, and teaching to the CCSS expectations; however, they acted as a willow in terms of the resources used in a lesson and the pace and order of topics addressed. 204 The flexibility in the district expectations requires a different monitoring system as well. The Lakeside reading instruction expectations, that allow for variation, would require any principal or outside observer walking the hallways at Cottonwood or Rose to not expect to see all teachers in a grade level to be teaching the same material, but instead these evaluators needed to carefully consider the lesson concepts addressed and the teacher and student actions throughout the lesson. Monitoring implementation of instructional strategies such as CAFE and Daily 5 required administrators to collect data on documents such as the HQI walkthrough forms that account for these actions, rather than simple checklists of if the teacher is teaching the “correct” lesson for that day and time. Overall when instructional strategies allow for the teacher to make modifications and to have variation across classrooms, both administrators and teachers must be thoughtful and reflective about the programs chosen, the implementation process, and monitoring systems. Implications for Researchers Like with school turnaround policies, this research adds to the research calling for policymakers to consider context when deciding what works. Currently policymakers and research funders prioritize studies using experimental science methods, such as randomized control trials, in search of programs that work in a decontextualized space. The leaders and teachers at Lakeside provide an alternative to considering which programs “work.” Instead policymakers should prioritize research into systems that provide guidance for teachers and administrators working to understand how to make modifications for local context. This research would require a departure from quantitative studies detached from people and circumstances. Instead future scaling up research should examine teacher decision-making around scaling up certain features of programs. 205 From a design standpoint, this study points to the importance that researchers consider that systems and school improvement efforts are dependent on the work of many stakeholders at multiple, embedded levels of the organization. Improvement science assumptions ask researchers to consider how context matters for the implementation of programs. In education, this context includes the classroom, the school, the district, and the community. In this study, I was present at meetings at both the district and school levels. I also observed how concepts discussed in these meetings penetrated the classroom environment. This data allowed me to trace how leaders’ orientations and the decisions made at each level of the organization had implications for the design and implementation of Blueprint systems. The design of this research could provide a framework for future studies seeking to consider organizations and systems. Future Research Future research can build on this research to further understand how leaders and embedded systems in a school organization influence implementation of school improvement and instructional reforms. First, future research could more deeply examine the affordances and constraints of placing certain decision-making responsibilities at different levels of the school organization. Are some instructional decisions better made at the classroom or school-level? In other words, what mandates are best determined at the district level that can be consistently implemented across school contexts? And what instructional policies are better established at a level closer to the teacher and students? Additionally, this study did not examine the effect of these programs on students. Implementing a new literacy instructional system at scale across a district impacts the students in the classrooms as much as the teachers. Consequently, future research could examine how 206 students experience instructional expectation changes in their learning environment. Is there any difference in experience between students in classrooms changing under each paradigm? Third, a quantitative study of the impacts of an improvement science system in which teachers are encouraged to make modifications would be insightful, since experimental science researchers argue that modifications away from fidelity reduce the effect found in RCT studies. How does mandating systems that build in opportunities for teachers to exercise some professional discretion impact how these systems affect student achievement? What is the relationship between a teacher’s expertise in the impacted content area or pedagogical content knowledge and student achievement when a program relying on professional discretion is mandated? Lastly, research arising from this dissertation could study how to train school administrators to engage with change under the improvement science paradigm. This study highlights that the improvement science paradigm will ask district and school leaders to think differently about their roles when leading school improvement efforts. Such research could study the systems for making decisions, considering modifications, and communicating expectations that could inform future professional learning opportunities to train administrators around these systems and the assumptions within them. Conclusions The goal of this research was to contribute to our understanding of school improvement and addressing the problem of scale. This research examined in what ways a school turnaround initiative based on improvement science principles worked to influence instructional practices at scale through embedded case studies of two elementary schools in two districts. By observing and interviewing teachers multiple times throughout the 2016-2017 school year, I was able to get 207 an understanding of what contributes to a teacher’s willingness and ability to change their instructional practice in alignment with a framework developed at the district level. Moreover, the school surveys of the entire staff allowed me to assess more broadly the formal and informal influences that supported a teacher’s effort to adjust their practice. This dissertation research also added to the literature by examining the ways that the work of improvement science reforms and networks can influence instructional practice at scale. Thus far, the research on NICs from the Carnegie Foundation has examined the use of NICs to change feedback cycles with teacher evaluations completed by the school principal (Bryk et al, 2015), student progression through college-level remedial math (Bryk et al., 2010), and developing interventions that support the development of a growth mindset in students (Carnegie Foundation, 2016). While all of these networks were working to address complex problems, this research extends the use of NICs to systems development and instructional improvement across grades and subjects throughout a district. Lastly, this research also has the potential to inform policy discussion at a time of transition in the school accountability policy landscape. With the passage of ESSA and the renewed flexibility it affords to state and district policymakers in determining how to improve persistently low-achieving schools, this dissertation provides research on an alternative to the external accountability practices advocated under NCLB and RTTT. This framework of using the expertise already in the district in a NIC-like system could lead to school improvement in a way that appreciates the expertise already established in an organization and still impacts the instructional core at scale throughout an entire low-performing district. 208 APPENDICES 209 APPENDIX A: Timeline of Blueprint Installation 210 APPENDIX B: Observation Protocol 211 APPENDIX C: Sample Vision of High Quality Reading Instruction The Vision of Ambitious Teaching of Reading Reading instruction has always been stressed for elementary school students, but today it takes on increased importance for all grades. Reports like Time to Act and Reading at Risk raise concern about a lack of depth in the literacy education of adolescent students and lament a general decline in reading among young adults. The [Michigan] State Standards for reading state that “all students must be able to comprehend texts of steadily increasing complexity as they progress through school,” and studies of literacy point to the rising expectations for reading in both schooling and the workplace. Documents like these indicate that teachers need to help all students become readers, regardless of whether they are in elementary or secondary school, so they can succeed in the information age. Because the curriculum in upper grades relies more and more on independent reading skills, students who struggle with reading in the early grades will eventually struggle in all academic areas. On average, 25% of schoolchildren in the early grades struggle with reading. Regardless of the causes of these students’ academic difficulty, research has demonstrated effective methods for providing them with early reading support. Two terms are circulating in current discussions of reading instruction: textual complexity and close reading. Textual complexity is defined in the [Michigan Standards] as a three-part entity. It includes quantitative dimensions such as word length or frequency, sentence length, and cohesion, all of which can be measured by computer software; qualitative dimensions such as levels of meaning, clarity of language, and knowledge demands, all of which require human readers; and reader-text variables such as reader motivation, knowledge, and experience, qualities best assessed by teachers who know students and texts. Both the qualitative dimensions and the reader-text variables depend upon the professional judgment of teachers, especially the reader-text variables, because only teachers know students well enough to help them find the best text for the purpose at hand, something “leveling” systems cannot do. Research on student readers and the texts they read confirms the need for teachers to play a key role in matching individual students with specific books at appropriate levels of textual complexity: What we know about our students as readers: • Students come to reading tasks with varied prior reading experiences, or prior knowledge, which can support their reading of complex texts. • Students who are engaged and motivated readers read more often and read more diverse texts than students who are unmotivated by the reading task. • Students who develop expertise with a particular kind of reading (science fiction or online games, for example) may not think their teachers will value this kind of reading. What we know about the texts students read: 212 • In and out of school, the texts students read vary significantly, from linear text-only books to multimodal textbooks to online hypertexts, each of which places different demands on readers and requires different strategies and approaches to reading. • Students read texts from a variety of disciplines, so content area literacy is important. (cid:1) • The level of difficulty or complexity in a text is not the only factor students consider in choosing texts; interest and motivation also matter. (cid:1) • Readability or levels can vary significantly within a single text, so it is important to consider other dimensions of textual complexity. Close reading has been proposed as the way to help students become effective readers of complex texts, and it can be useful, especially when used alongside other approaches. The difficulty is that close reading can be defined in multiple terms. It can mean searching for hidden meanings, positioning the text as the only reality to be considered, and focusing on formal features. Close reading is also a highly contested term among college English instructors. Critics condemn it for conceptualizing the text as a closed world, for limiting student access, and for emphasizing form over content. Furthermore, research shows that reading comprehension depends on a more complex approach. Specifically, reading comprehension results from the integration of two models, text- based and situation-based. The text-based model focuses on the way words are organized into sentences, paragraphs, and whole texts. The situation model refers to the meaning that results from integration of the text-based approach with the reader’s prior knowledge and goals. Close reading is aligned with the text-based approach, and it encourages students to see meaning as one right answer to be extracted from the text. Close reading is often conflated with providing textual evidence for making a claim about a text, but any approach to reading can insist on warrants for interpretations of texts. By itself, then, close reading cannot ensure that students will develop deep understandings of what they read. A good reading comprehension program includes a systematic approach to introducing vocabulary and background information, specific strategies with cumulative review and practice, and the integration of the skills during passage reading (Jitendra & Gardill, 1994). Reading comprehension skills improve when teachers systematically guide students to attend to story elements and model the thought process behind each strategy (Bos & Vaughn, 1998; Mathes, Fuchs, & Fuchs, 1997; Taylor, Harris, Pearson, & Garcia, 1995). Instructional Implications Research-based understandings about students, texts, and reading underlie instructional approaches that support students’ learning to read complex texts across grade levels and disciplines. NCTE and the National Reading Panel affirm the value of multiple approaches and support teachers’ efforts to adopt instructional practices that call upon a variety of effective strategies including the following: • Explicit instruction of the five essential components of effective reading instruction: 213 phonemic awareness, phonics, comprehension, fluency, and vocabulary. • Recognize the role that motivation plays in students’ reading by modeling for students how to engage with complex texts that do and do not interest them. • Engage students in performative reading responses such as gesture, mime, vocal intonation, characterization, and dramatization to enable active construction of meaning and construct a collaborative environment that builds on the strengths of individual students. • Have students read multiple texts focused on the same topic to improve comprehension through text- to-text connections. • Foster students’ engagement with complex texts by teaching students how different textual purposes, genres, and modes require different strategies for reading. • Encourage students to choose texts, including non- fiction, for themselves, in addition to assigned ones, to help them see themselves as capable readers who can independently use reading capabilities they learn in class. • Demonstrate, especially at the secondary level, how digital and visual texts including multimodal and multi-genre texts require different approaches to reading. • Connect students’ reading of complex texts with their writing about reading and with writing that uses complex texts as models so they will recognize and be able to negotiate many different types of complex texts. • Develop students’ ability to engage in meaningful discussion of the complex texts they read in whole- class, small group, and partner conversations so they can learn to negotiate and comprehend complex texts independently. When teachers can choose from a range of research-based and theoretically grounded instructional approaches, their students learn how to choose from, apply, and reflect on diverse strategies as they take up the varied purposes, subjects, and genres that present complex challenges for readers. Formative assessment of reading can take many forms, as the examples below show: • Teachers can help students develop awareness of their diverse experiences and knowledge—all of which affect the ways they engage with texts. These include reading experiences in previous grades and in out-of-school spaces. Once students have identified their experiences and knowledge, teachers can help students build on them in approaching complex texts—including when their background experiences and knowledge enhance and/or interfere with their ability to read complex texts. • Asking students to think aloud as they read complex texts can help teachers identify which instructional supports and interventions will best support readers as they face new reading challenges. • When teachers have identified students who struggle to remain engaged as they read complex texts, they can assess students’ interests in order to provide texts that are more likely to foster student engagement. • Teachers can assess students’ ability to think about their reading and about how different kinds of texts impact their reading. This increased awareness can improve students’ ability to read complex texts for various purposes. 214 APPENDIX D: Fall Superintendent Interview Protocol 1. Please tell me about your background in education. a. What administrative positions have you held? b. How long have you been in your current position? c. What drew you to work in this district? 2. Please tell me about your district. a. Students? b. Teachers? c. Longevity? d. Community? 3. What are your goals for your district this year? 4. How did you learn about the Blueprint for Systemic Reconfiguration? a. How was it framed to you? b. What were your initial thoughts? 5. Could you describe your district’s progress with the Blueprint for Systemic Reconfiguration? a. When did you start? b. What is your timeline? c. How was it made? d. What elements have you implemented? When? e. Why did you choose to begin with these elements? f. How is each listed element going? Challenges? Adjustments made? g. How are you measuring the progress of each component? 6. Can you describe the current district plan to influence the teacher instructional practices in English Language Arts? a. How did you develop the plan? b. Why did you develop it in this way? c. How is this plan being communicated to the teachers? d. How will they work to change their practice? e. Where have you invested resources toward ELA instruction? f. People? g. Why are these elements the focal points for the building? h. What supports will be given to teachers? i. What obstacles and/or challenges do you anticipate? j. How are you thinking about addressing these? 7. What informed how you developed the plan for ELA instruction? a. Any particular curriculum? b. What resources did you use to design this plan? 215 8. With unlimited time and resources, how would this plan look different? 9. How did you choose people to join the DN? a. What criteria did you have? b. Why did you invite them? c. How did you invite them? d. How did you describe (frame) the DN to them? 10. How does your DN work? a. How are decisions made in the DN? b. How do the BNs compare to the DN? c. run similarly or differently 11. What is your role in supporting the work of the Blueprint? a. At the district level? At the building level? b. (probe for specifics in influencing changes in the Reading instructional core) c. How often are you in classrooms? d. How will your role evolve during the school year? e. How many hours in a day do you (can you) commit to the Blueprint? f. What does this work look like? 12. What contact do you have with others outside of your district to support your work regarding the Blueprint? a. How do they influence/guide your work? b. What is your relationship with this person? c. How often do you meet with other principals about this work? 13. How do you distribute information from the DN to the BN? a. How is it shared through your school? b. How should it be used by the BN? 14. Alternatively, how does information from the BNs get to the BN? a. What kind of information do you hope to get? 15. I know that recently the state of Michigan has allocated quite a bit of funding toward reading, have you received any of these funds? a. If so, what for? b. How does this funding work with your district’s work in improving Reading instruction as part of the Blueprint? 216 APPENDIX E: Spring Superintendent Interview Protocol 1. Reflecting back on the year, what were some of the positives? a. Anything with the Blueprint in particular b. Literacy instruction 2. What were some of your biggest challenges? a. Anything with the Blueprint in particular b. Literacy Instruction DN? 3. What elements of the Blueprint were implemented in your district this year? a. How did it go? b. Anything unexpected happen with implementation? 4. How did you share information from the DN to the schools? a. How did you get information from the schools to the DN? i. Can you share any examples of something from the schools informing the 5. How did you decide what topics to cover each week at DN? 6. Did you complete any classroom observation walkthroughs? a. How were you trained to complete these observations? b. How did it feel to complete these? c. How did you give feedback? d. How did you decide which classrooms to visit during HQ walkthroughs? e. How did teachers respond? 7. Another thing that I noticed this year is a focus on data. Can you tell me the ways you used data this year? a. How did it inform your work? b. How was this different than from in past years? 8. Mentions of union pushback. Would you be willing to share some of what happened this year? 9. What are the next steps for the Blueprint for next year? 10. What changes are you planning on making for next year? a. Probe for literacy instruction b. Probe for professional development c. Probe for superintendent role d. Probe for around the school 11. Overall this year what did the Blueprint mean for your work and the district? 217 APPENDIX F: Fall Building Principal Interview Protocol 1. Please tell me about your background in education? a. What administrative positions have you held? b. How long have you been in your current position? c. What drew you to work in this district? 2. Please tell me about your school. a. Students? b. Teachers? c. Longevity? d. Community? 3. What are your goals for your school this year? 4. How did you learn about the Blueprint for Systemic Reconfiguration? a. How was it framed to you? b. What were your initial thoughts? 5. Could you describe your building’s progress with the Blueprint for Systemic Reconfiguration? a. When did you start? b. What is your timeline? c. How was it made? d. What elements have you implemented? When? e. Why did you choose to begin with these elements? f. How is each listed element going? Challenges? Adjustments made? g. How are you measuring the progress of each component? 6. Can you describe the current BN plan to influence the teacher instructional practices in English Language Arts? a. How did you develop the plan? b. Why did you develop it in this way? c. How is this plan being communicated to the teachers? d. How will they work to change their practice? e. Where have you invested resources toward Reading instruction? f. People? g. Why are these elements the focal points for the building? h. What supports will be given to teachers? i. What obstacles and/or challenges do you anticipate? j. How are you thinking about addressing these? 7. What informed how you developed the plan for ELA instruction? a. Any particular curriculum? b. What resources did you use to design this plan? 218 8. With unlimited time and resources, how would this plan look different? 9. How did you choose people to join the BN? a. What criteria did you have? b. Why did you invite them? c. How did you invite them? d. How did you describe (frame) the BN to them? 10. How does your BN work? a. How are decisions made in the BN? b. How does the BN compare to the DN? c. Do they run similarly or differently 11. What is your role in supporting the work of the Blueprint? a. At the district level? At the building level? b. (probe for specifics in influencing changes in the Reading instructional core) c. How often are you in classrooms? d. How will your role evolve during the school year? e. How many hours in a day do you (can you) commit to the Blueprint? f. What does this work look like? 12. What contact do you have with others outside of your school to support your work regarding the Blueprint? a. How do they influence/guide your work? b. What is your relationship with this person? c. How often do you meet with other principals about this work? 13. How do you receive information from the DN? a. How is it shared through your school? b. How is it used by the BN? 219 APPENDIX G: Spring Building Principal Interview Protocol 1. Reflecting back on the year, what were some of the positives from the year? a. Anything with the Blueprint in particular b. Literacy instruction 2. What were some of your biggest challenges? a. Anything with the Blueprint in particular b. Literacy instruction 3. What elements of the Blueprint were implemented in your school this year? a. How did it go? b. Anything unexpected happen with implementation? 4. I noticed this year, an emphasis on the HQ observations can you tell me about that process? a. How were you trained to complete these observations? b. How did it feel to complete these? c. How did you give feedback? d. How did teachers respond 5. Another thing that I noticed this year is a focus on data. Can you tell me the ways you used data this year? a. How did it inform your work? b. How was this different than from in past years? 6. How did you get information from the DN? a. How did you spread that information around your school? b. Spread information from the BN around your school? 9. Can you talk a little bit about the professional development that you did at your school around the Blueprint (HQ instruction / HQ walk-throughs)? a. How did you come up with those ideas? 7. Roles and Responsibilities at BN a. How did that begin? 8. CAFÉ strategies a. What supports did you give teachers? 10. What are your plans for next year? a. Probe for job placement b. Probe for changes in instruction c. Probe for new roles in the school 220 11. What changes are planning on making for next year? a. Probe for literacy instruction b. Probe for professional development c. Probe for grade level d. Probe for around the school 221 APPENDIX H: Teacher Interview Protocol – Round 1 1. Please tell me about your background in education? a. How long have you been in your position? b. What drew you to work in this district? 2. Please tell me about your school and classroom. a. Students? b. Other Teachers? c. Community? 3. What are your goals for your ELA instruction this year? 4. Can you walk me through how you planned this lesson and please try to be as specific as possible? a. How did you start? b. When did you do the planning of today’s lesson? c. How did you decide what content to focus on? d. How did you decide on the structure of your lesson? e. What resources did you use when you planned? 5. Did you plan today’s lesson with any other teachers? a. How did you plan together? b. When did you plan the lesson? 6. Was there anything different in how you planned this lesson from usual? 7. Are there any other supports that you have for your ELA instruction? a. How do these materials support you? b. Why do you find it helpful? 8. What professional development experiences focused on ELA have you had since the beginning of the year? a. What was the focus? b. Who facilitated it? 9. What information have you gotten on the Blueprint since the beginning of the year? a. What do you think of this information? b. Is it helpful? c. Has it influenced your work in any way? (in the classroom? Outside the classroom?) (listen for coaches) d. What do your colleagues think about this information? e. Are there any key colleagues that you go to for information on the Blueprint? f. More specifically Reading instruction according to the Blueprint? g. Have you spoken with anyone outside of your school about the Blueprint? 222 APPENDIX I: Teacher Interview Protocol – Round 2 1. Can you walk me through how you planned this lesson and please try to be as specific as possible? a. How did you start? b. When did you do the planning of today’s lesson? c. How did you decide what content to focus on? d. How did you decide on the structure of your lesson? e. What resources did you use when you planned? 2. Before the changes over the past couple of years, in what ways would this lesson would have looked different? 3. Did you plan today’s lesson with any other teachers? 4. Was there anything different in how you planned this lesson from usual? 5. Have you had any additional professional learning opportunities since November on Reading? 6. I last saw you in November, can you share with me what experiences you have had with the Blueprint since the last time we talked? a. Last time we talked you discussed how this Blueprint work has promoted more collaboration. Is this still true? b. Now that it is two months later, how are teachers feeling with all of this change? 223 APPENDIX J: Teacher Interview Protocol – Round 3 1. Can you walk me through how you planned this lesson and please try to be as specific as possible? a. How did you start? b. When did you do the planning of today’s lesson? c. How did you decide what content to focus on? d. How did you decide on the structure of your lesson? e. What resources did you use when you planned? 2. Before the changes over the past couple of years, in what ways would this lesson would have looked different? 3. Now that we’re through the first semester - When you think about your students, have you seen differences in your students’ learning or behavior from previous years. 4. Did you plan today’s lesson with any other teachers? 5. Was there anything different in how you planned this lesson from usual? 6. Have you had any additional professional learning opportunities since November on Reading? 7. I have learned and heard from some teachers over the past few months that they are getting observed a lot more this year. Can you tell me about that process as a teacher? a. What are the observers looking for? b. What kind of feedback are you getting? c. How does it feel as a teacher? Has that feeling changed over the school year? 8. Another thing that I have heard being mentioned a lot is data. Can you tell me about how you are using data this year? a. How is the data informing what happens in your classroom? 9. What is your stress level like? a. Why? 10. I last saw you in January, can you share with me what experiences you have had with the Blueprint since the last time we talked? 224 APPENDIX K: Teacher Interview Protocol – Round 4 1. Can you walk me through how you planned this lesson and please try to be as specific as possible? a. How did you start? b. When did you do the planning of today’s lesson? c. How did you decide what content to focus on? d. How did you decide on the structure of your lesson? e. What resources did you use when you planned? 2. Did you plan today’s lesson with any other teachers? 3. Before the changes over the past couple of years, in what ways would this lesson would have looked different? 4. Have you had any additional professional learning opportunities since April in literacy? 5. Overall, how have your teaching practices in literacy changed this year from previous years? a. Any new routines and teaching practices? b. Any routines and/or teaching practices that you have stopped doing? 6. Reflecting back on the year, what were some of the positives from the year? a. Probe for literacy instruction 7. What were some of your biggest challenges? a. Probe for literacy instruction 8. If you were in charge, what changes would you make for next year? a. Probe for literacy instruction b. Probe for professional development c. Probe for grade level d. Probe for around the school 9. Since I last saw you in April, can you share with me what experiences you have had with 10. Now thinking about the Blueprint, in particular, how has this program changed your the Blueprint? instruction? a. Work outside of the classroom? 11. What are your plans for next year? a. Probe for job placement b. Probe for changes in instruction c. Probe for new roles in the school 225 APPENDIX L: Fall Teacher Survey 226 227 228 229 230 231 232 233 234 235 APPENDIX M: Spring Teacher Survey 236 237 238 239 240 241 242 243 244 245 246 APPENDIX N: Theory of Change for the Blueprint for Systemic Reconfiguration 247 APPENDIX O: Sample Instructional Infrastructure Planning Tool – Floor 1 248 APPENDIX P: Lakeside School District HQI Walkthrough Observation Form 249 APPENDIX Q: Hillside Curriculum Framework- Unit 2 Stage 1: Identify Desired Results Essential Question: How do we pursue our passions and why is this pursuit important to our quality of life? Scaffold Questions: How do we define passion? Brief Summary of Unit: How might our natural talents strengthen our passions? How do our unique talents shape our passions? Why is it important to be true to one’s passions? What does it mean to persevere? How can persevering your passions improve your life? How can change affect one’s passion? How can the pursuit of our passions impact others? Through the use of read-alouds, video clips, and leveled texts, students will be able to define passion. Students will understand how pursuing one’s passion can lead to a satisfying and rewarding life. Through the use of read-alouds, video clips, and leveled texts, students will be able to define passion and provide examples of characters and real-life people who have pursued their passions. They will listen, read, write and discuss how pursuing one’s passion can improve one’s quality of life. Teachers will progress monitor throughout the unit using running records, checklists, rubrics, etc. Teachers will be provided with materials and training. Weeks 1-6: Read aloud and have accountable talk with The One and Only Ivan by Katherine Applegate, across the 5 week unit. As you read aloud, discuss how Ruby’s arrival awakens something in Ivan and helps him realize his passions and how the two pursue their passion. (Note: This chapter book is an additional resource and is not included in the Journey’s resources.) Week 1: This week the class will explore the following scaffold questions: How do we define passion? How might our 250 natural talents strengthen our passions? The following resources will provide the class opportunities to read, write, and discuss passion: Bat Loves the Night - (anchor text) Passion Quotes - to inspire thoughtful conversations enabling students to more clearly understand the concept of passion. The Boy Who Loved Math: The Improbable Life of Paul Erdos This picture book tells the story of Paul Erdos who, at the age of four, would ask individuals when they were born and then calculate the number of seconds you had been alive in his head. But he didn't learn to butter his own bread until he turned twenty. Week 2: This week the class will explore the following scaffold question: How do our unique talents shape our passions? Why is it important to be true to one’s passions? The following resources will provide the class opportunities to read, write, and discuss the importance of being true to ourselves and our unique passions: What Do Illustrators Do? (anchor text) The Boy Who Drew Birds: A Story of John James Audubon by Jacqueline Davies - This picture book tells the story of James Audubon and how his passion for bird watching as a young boy led him to become America’s greatest painter of birds. Interview with illustrator, Dan Yaccarino - the illustrator discusses how his passion for art goes back as far as he can remember and how that passion Week 3: This week the class will explore the following scaffold questions: What does it mean to persevere? How can persevering your passions improve your life? The following resources will provide the class opportunities to read, write, and discuss the importance persevering one’s passion. The Harvest Birds (anchor text) Emmanuel’s Dream-The True Story of Emmanuel Ofosu Yeboah by Laurie Ann Thompson - This picture book tells the true story of a young boy who perseveres his passions and triumphs over adversity. Emmanuel becomes a champion for the disabled and continues spreading his powerful message: disability is not inability Trailer to the Movie: Emmanuel's Dream Explanation of Grit Believe in Yourself - Never Give Up Video 251 Week 4: This week the class will explore the following scaffold questions: How can change affect one’s passions? The following resources will provide the class opportunities to read, write, and discuss how change can impact the pursuit of one’s passions. Kamishiba Man (anchor text) The True Story of Kamishibai (paired informational text found on page T334) Whatever it Takes: Paralyzed Motocross Rider Darius Glover (4:50 min) an inspiring story of a promising motocross rider who’s life was changed forever when he was paralyzed in a motocross accident. This story can also be related to last week’s theme of perseverance and grit. Week 5: This week the class will explore the following scaffold questions: How do ? The following resources will provide the class opportunities to read, write, and discuss how can the pursuit of our passions impact others. Young Thomas Edison (anchor text) Moving Pictures (paired informational text One Plastic Bag: Isatou Ceesay and the Recycling Women of the Gambia by Miranda Paul; This book tells the story of Ceesay who saw that people were tossing used plastic bags on the ground just as they had always thrown away their baskets when no longer useful—except the plastic bags, unlike the baskets, weren't biodegradable. So Ceesay figured out how to use crochet, a skill with which the villagers were already familiar, to make purses out of the plastic bags. How Little People Can Make a Big Difference (4:47 min video) This video demonstrates how Charlie Cooper a young, friendless boy became passionate about finding a solution to his friendship problem. His idea had a positive impact on many students across Australia. Make a Difference ( 1:00 minute Pass It On video) A video to prompt discussion. Who is passionate in this video? The teacher, the little girl, the mother? Why do you think that? What are their passions? How did one character’s passion impact the others? Students will understand how pursuing our passions can enrich our lives as well as the lives of those around us. Sometime pursuing our passions can impact the lives of people we don’t 252 Desired Understanding: Key Knowledge and Skills (Content): even know. Students will understand that sometimes our unique talents encourage and strengthen our passions. Students will also understand how one’s passion can help an individual overcome an adversity. Students will also understand that pursuing our passions can have an affect on other people as well. Week 1: Our natural talents can ignite and strengthen our passions. Week 2: Pursuing our unique talents can lead to fulfilling careers and enrich our lives. Week 3: Many times throughout our lives we may experience adversity. How can pursuing one’s passions help us to overcome adversity and live rich lives. Week 4: Change can affect our ability to pursue our passions. Week 5: An individual’s passion can positively impact other people’s lives. Reading Skills Identify a sequence of events Asking and answering questions about key details Determine the meaning of domain specific vocabulary Using text and graphic features to enhance comprehension Analyze and evaluate text as you read Drawing conclusions Literal and nonliteral meanings Inferring Predicting Cause and Effect Analyze Illustrations Main Idea and Details Summarize Language Skills Identify action verbs Identify and use being verbs Identify complete subjects and complete predicates Produce and use sentences with compound predicates Identify the past, the present, and the future tense of verbs Identify abstract nouns Identify pronouns and antecedents Writing Skills Response paragraph 253 Common Core State Standards (CCSS) - Reading Common Core State Standards (CCSS) - Writing ○ RL.3.1, RI.3.7, RL.3.4, RL.3.5, W.3.3b, W.3.5, Opinion writing Purpose Organization Supporting opinions with strong reasons Elaboration Conventions Word Study Noticing the Vcv pattern helps you decide where the wor should be divided Spelling words with long and short vowel sounds Suffixes: -able, -ible Three letter clusters Silent letters: kn, wr Diphthongs: ow and ou Words with au, aw, al and o RI.3.4. Determine the meaning of general academic and domain- specific words and phrases in a text relevant to a grade 3 topic or subject area. RI.3.3. Describe the relationship between a series of historical events, scientific ideas or concepts, or steps in technical procedures in a text, using language that pertains to time, sequence, and cause/effect. RL.3.1. Ask and answer questions to demonstrate understanding of a text, referring explicitly to the text as the basis for the answers. RL.3.5. Refer to parts of stories, dramas, and poems when writing or speaking about a text, using terms such as chapter, scene, and stanza; describe how each successive part builds on earlier sections. RL.3.7. Explain how specific aspects of a text’s illustrations contribute to what is conveyed by the words in a story (e.g., create mood, emphasize aspects of a character or setting). W.3.1. Write opinion pieces on topics or texts, supporting a point of view with reasons. W.3.4. With guidance and support from adults, produce writing in which the development and organization are appropriate to task and purpose. W.3.5 With guidance and support from adults, produce writing in which the development and organization are appropriate to task and purpose. 254 Common Core State Standards (CCSS) - Listening & Speaking Common Core State Standards (CCSS) - Language W.3.8. Recall information from experiences or gather information from print and digital sources; take brief notes on sources and sort evidence into provided categories. SL.3.1. Engage effectively in a range of collaborative discussions (one-on-one, in groups, and teacher-led) with diverse partners on grade 3 topics and texts, building on others’ ideas and expressing their own clearly. SL.3.2. Determine the main ideas and supporting details of a text read aloud or information presented in diverse media and formats, including visually, quantitatively, and orally. L.3.1. Demonstrate command of the conventions of standard English grammar and usage when writing or speaking. a. Explain the function of nouns, pronouns, verbs, adjectives, and adverbs in general and their functions in particular sentences. c. Use abstract nouns (e.g., childhood) L.3.2c Demonstrate command of the conventions of standard English capitalization, punctuation, and spelling when writing. c. Use commas and quotation marks in dialogue. L.3.4. Determine or clarify the meaning of unknown and multiple-meaning word and phrases based on grade 3 reading and content, choosing flexibly from a range of strategies. b. Determine the meaning of the new word formed when a known affix is added to a known word (e.g., agreeable/disagreeable, comfortable/uncomfortable, care/careless, heat/preheat). c. Use glossaries or beginning dictionaries, both print and digital, to determine or clarify the precise meaning of keywords and phrases. L.3.5 Demonstrate understanding of figurative language, word relationships and nuances in word meanings. c. Use a known root word as a clue to the meaning of an unknown word with the same root (e.g., company, companion). L.3.6. Acquire and use accurately grade-appropriate conversational, general academic, and domain-specific words and phrases, including those that signal spatial and temporal relationships (e.g., After dinner that night we went looking for them). 255 Common Core State Standards (CCSS) – Foundational Skills Alignment to the Vision of High Quality Instruction in Reading & Writing RF.2.3. Know and apply grade-level phonics and word analysis skills in decoding words. a. Distinguish long and short vowels when reading regularly spelled one-syllable words. b. Know spelling-sound correspondences for additional common vowel teams. d. Decode words with common prefixes and suffixes. e. Identify words with inconsistent but common spelling-sound correspondences. RF.3.3.c Know and apply grade-level phonics and word analysis skills in decoding words.: Decode multi-syllable words RF.3.3 Read with sufficient accuracy and fluency to support comprehension. Reading ● Explicit instruction of the five essential components of effective reading instruction: phonemic awareness, phonics, fluency, and vocabulary, and comprehension. ● Foster’s students’ engagement with complex texts by teaching students how different textual purposes, genres, and modes require different strategies for reading. ● Develop students’ ability to engage in meaningful discussion of the complex texts they read in whole-class, small group, and partner conversations so they can learn to negotiate and comprehend complex text independently. ● Connect students’ reading of complex texts with their writing about reading and with writing that uses complex texts as models so they will recognize and be able to negotiate many different types of complex texts. ● Formative assessment Writing ● Authentic writing takes place in a real-world context and addresses real-world context and addresses real-world needs. ● Require all students to write extensively so that they can be comfortable writing extended prose in elementary school. ● Explicit teaching of grammar using a context-based ● Make new-media (technology) writing part of students’ functional approach. regular composing. ● Use strategies of formative assessment to give students feedback on developing drafts. ● Employ multiple measures, including portfolios, to provide summative assessments of students’ progress. 256 Stage 2: Determine Acceptable Evidence Performance Task (Competence-Centered) Teachers will create weekly anchor charts to record students’ thinking and ideas in response to the anchor text as well as the other weekly resources. Students will create a Unit 2 - Pursuing Your Passions Portfolio. Students will reflect on their readings each week and answer the weekly essential question. These answers should include evidence from the weekly texts. ( RL.3.5, W.3.2, W.3.4 L.3.3.a, L.3.6) ● Students will respond to the text we read aloud throughout the unit, The One the Only Ivan. Students will complete the Unit 2 Assessment answering the essential question: How do we pursue our passions and why is this pursuit important to our quality of life? Use text evidence from your own experience and The One the Only Ivan to support your opinion. Degree of Competence/Criteria Other Evidence (Content-Centered) UNDERLINED ASSESSMENTS ARE REQUIRED. Additional assessments are listed as suggestions for progress monitoring. ● Teachers will administer weekly tests 1 and 4 located in the Grab-and-Go! Resource Kit. These weekly tests 257 Student Self- Reflection and Self- Regulation (Student-Centered) System-Centered Competence assess students’ knowledge of comprehension, vocabulary, and phonics skills taught in Unit 1. ○ Unit 2-Lesson 1 Test: A Bat Loves the NIght ■ RL.3.3, RL.3.2, L.3.6, L.3.4a, RF.3.3c, L.3.1i, L.3.1h ○ Unit 2- Lesson 4 Test: Kamishiba Man ■ L.3.6, L.3.4c, RL.3.3, RF.3.3c, L.3.1a, L.3.2a ● Students will be given differentiated word study lists. Teachers will assess the students at the beginning and end of the year. Teachers will administer the Qualitative Spelling Inventory (QSI) found on p. 395 in the Literacy and Language Guide to inform them what word study list will be appropriate for each groups of students. NOTE: Teachers will progress monitor throughout the unit using running records, checklists, rubrics, etc. ● Retell the anchor text. Use the prompts on the retelling cards to guide children to retell the story or guide children to identify the main topic and retell key details of a text. Use the rubric included in the Teacher Guide as an assessment tool. (RI.3.2, RL.3.2) ● Turn and Talk about the weekly essential question during read-alouds and Use the Collaborative Conversations rubric R15 to assess students.( SL. 3.1a, SL.3.1b, SL.3.1c, SL.3.1d) ● Student Self Assessment (in folder) ● Student Reading Goal Page (in folder) ● Students will complete 2 sections of the Journeys Unit 2 Benchmark Assessment on their Chromebooks. The assessment can be found online under Program Assessments in Benchmark and Unit Tests. ○ Treasures Found questions 1-18 258 ○ Pine Elementary School Science Fair only question 36 (Written response only) ○ Birthday Chickens questions 38-42 (Revising and Editing) ○ Standards assessed: RF.2.3.b, RF.2.3.d, RF.2.3.e, L.3.2c, L.3.1c, L.3.4b, L.3.5c L.3.4c, L.3.6, RL.3.1, RI.3.7, RL.3.4, RL.3.5, W.3.3b, W.3.5, W.3.8 259 APPENDIX R: Lakeside School District High Quality Instruction Implementation Guide 260 APPENDIX S: CAFE Menu Bulletin Board (from The Daily Cafe, 2018) 261 APPENDIX T: CAFE Link in Lakeside Implementation Guide 262 APPENDIX U: Cottonwood May PSP Form 263 APPENDIX V: Sample Mandated Anchor Chart 264 REFERENCES 265 REFERENCES Adler, P. S., & Kwon, S. W. (2002). Social capital: Prospects for a new concept. Academy of Management Review, 27(1), 17-40. Anrig, G. (2015, July 25). Lessons from School Improvement Grants that Work. The Century Foundation. Argyris, C., & Schön, D. A. (1996). Organizational Learning II: Theory. Method and Practice. Atteberry, A. & Bryk, A.S. (2010). Centrality, connection, and commitment: The role of social networks in school-based literacy. In A. J. Daly (Ed.), Social network theory and educational change (pp. 51-76). Cambridge, MA: Harvard Educational Press. Baker-Doyle, K. J., & Yoon, S. A. (2010). Making expertise transparent: Using technology to strengthen social networks in teacher professional development. In A. J. Daly (Ed.), Social network theory and educational change (pp. 115-126). Cambridge, MA: Harvard Educational Press. Baxter, P., & Jack, S. (2008). Qualitative case study methodology: Study design and implementation for novice researchers. The Qualitative Report, 13(4), 544-559. Booher-Jennings, J. (2005). Below the bubble: “Educational triage” and the Texas accountability system. American Educational Research Journal, 42(2), 231-268. Boreham, N., & Morgan, C. (2004). A sociocultural analysis of organisational learning. Oxford Review of Education, 30(3), 307-325. Birckmayer, J. D., & Weiss, C. H. (2000). Theory-Based Evaluation in Practice What Do We Learn?. Evaluation Review, 24(4), 407-431. Brockway, P. (2017, August 2). Installing a Problem-Solving Driver System [PowerPoint Slides]. Retrieved from https://docs.google.com/presentation/d/1fC3IzkGu0rPgOcrUSatosFLQdS2vVz6Rn6bmO jrpDU8/edit#slide=id.p8. Bryk, A. S., Gomez, L. M., & Grunow, A. (2011). Getting ideas into action: Building networked improvement communities in education. In M. T. Hallinan (Ed.), Frontiers in sociology of education (pp. 127-162). New York: Springer. Bryk, A. S., Gomez, L. M., Grunow, A., & LeMahieu, P. G. (2015). Learning to Improve: How America's Schools Can Get Better at Getting Better. Cambridge, MA: Harvard University Press. 266 Carolan, B. V. (2013). Social network analysis and education: Theory, methods & applications. Los Angeles, CA: Sage Publications. Chandler, G. (2016). Blueprint for Turnaround. https://sites.google.com/a/calhounisd.org/mi- excel/home/blueprint-for-turnaround Chandler, G.A. (2017). Measuring Readiness to Install the Blueprint [PDF document]. Retrieved from https://miexcelresourcecenter.org/wp- content/uploads/2017/12/Measuring-District-Readiness-to-Install-the-Blueprint.pdf. Chandler, G.A. (2017). The Blueprint for Systemic Reconfiguration as an Evidence-Based Intervention [PDF document]. Retrieved from https://blueprinttoolsandresources.com/wp-content/uploads/2017/12/03-The-Blueprint- as-an-Evidence-Based-Intervention.pdf. Chandler, G.A. & Frank, J. (2015). The Blueprint for Systemic Reconfiguration [PowerPoint slides]. Retrieved from https://blueprinttoolsandresources.com/wp- content/uploads/2017/12/The-Blueprint-for-Systemic-Reconfiguration.pdf. Chandler, G. A. & Frank, J. (2017). The Blueprint’s Theory of Action [PDF document]. Retrieved from https://blueprinttoolsandresources.com/wp-content/uploads/2017/12/01- BP-Theory-of-Action-FINAL.pdf. Chander, G.A., Mohney, K.L., & Brophy, E.R. (2017). An Overview of the Blueprint’s Research Foundation [PowerPoint slides]. Retrieved from https://blueprinttoolsandresources.com/wp-content/uploads/2017/12/Overview-of-the- Blueprints-Research-Foundation.pdf. organizational knowledge sharing. Information & Management, 45(7), 458-465. Chow, W. S., & Chan, L. S. (2008). Social network, social trust and shared goals in Coburn, C. E., Mata, W. S., & Choi, L. (2013). The embeddedness of teachers’ social networks evidence from a study of mathematics reform. Sociology of Education, 86(4), 311-342. Coburn, C. E., & Russell, J. L. (2008). District policy and teachers’ social networks. Educational Evaluation and Policy Analysis, 30(3), 203-235. Cohen, D. K. & Ball, D.L.(1999). Instruction, capacity, and improvement. Philadelphia, PA: Consortium for Policy Research in Education, University of Pennsylvania. Cole, R. P., & Weinbaum, E. H. (2010). Changes in attitude: Peer influence in high school reform. In A. J. Daly (Ed.), Social network theory and educational change (pp. 77-96). Cambridge, MA: Harvard Educational Press. Coleman, J. S. (2000). Social capital in the creation of human capital. In E. L. Lesser (Ed.) Knowledge and social capital (pp. 17-41). Routledge. 267 Review, 66(1), 1-27. Elmore, R. F. (2004). School Reform from the Inside Out: Policy, Practice, and Performance. Harvard Educational Pub Group. Elmore, R. (2008). Improving the instructional core. Draft Manuscript. Elmore, R. F. (2016). “Getting to scale…” it seemed like a good idea at the time. Journal of Educational Change, 17(4), 529-537. Crossan, M. M., Lane, H. W., & White, R. E. (1999). An organizational learning framework: From intuition to institution. Academy of Management Review, 24(3), 522-537. Cross, R., Borgatti, S. P., & Parker, A. (2002). Making invisible work visible: Using social network analysis to support strategic collaboration. California Management Review, 44(2), 25-46. Cucchiara, M. B., Rooney, E., & Robertson-Kraft, C. (2015). “I’ve Never Seen People Work So Hard!” Teachers’ Working Conditions in the Early Stages of School Turnaround. Urban Education, 50(3), 259-287. Daly, A. J., Moolenaar, N.M, Der-Martirosian, C., Canrinus, E.T., & Chrispeels, J.H (s022), Der- merstioen, C., & Moolenaar, N. M. (2011). A capital investment: The effects of teacher human and social capital on student achievement in improving schools. In 24th International Congress on School Effectiveness and Improvement 2011:" Lining Research, Policy and Practice to promote Quality in Education". Daly, A. J., Moolenaar, N. M., Bolivar, J. M., & Burke, P. (2010). Relationships in reform: The role of teachers' social networks. Journal of Educational Administration, 48(3), 359-391. Daly, A. J., Moolenaar, N. M., Der-Martirosian, C., & Liou, Y. (2014). Accessing capital resources: Investigating the effects of teacher human and social capital on student achievement. Teachers College Record, 116(7). Dee, T. (2012). School turnarounds: Evidence from the 2009 Stimulus (No. w17990). National Bureau of Economic Research. Donaldson, M. L., & Papay, J. P. (2015). Teacher evaluation for accountability and development. In H. F. Ladd & E.B. Fiske (Eds.) Handbook of Research in Education Finance and Policy (pp. 174-193). Routledge. Duke, D. L. (2012). Tinkering and turnarounds: Understanding the contemporary campaign to improve low-performing schools. Journal of Education for Students Placed at Risk (JESPAR), 17(1-2), 9-24. Elmore, R. (1996). Getting to scale with good educational practice. Harvard Educational 268 English, P. & Robinson, R. (2017, August 1). Installing the Blueprint Networks [PowerPoint slides]. Retrieved from https://docs.google.com/presentation/d/1_xHUE5xNEBolQCEA8pAQAM1oTEFH90qlw wSD6j1hhlU/edit#slide=id.p7. Figlio, D. N., & Getzler, L. S. (2002). Accountability, Ability and Disability: Gaming the System (No. w9307). National Bureau of Economic Research. Finnigan, K. S., & Daly, A. J. (2012). Mind the gap: Organizational learning and improvement in an underperforming urban system. American Journal of Education, 119(1), 41-71. Finnigan, K.S. & Daly, A.J. (2016). Why We Need to Think Systemically in Educational Policy and Reform. In A.J. Daly & K.S. Finnigan (Eds.), Thinking and Acting Systemically: Improving School Districts Under Pressure (pp.11-48). Washington DC: American Educational Research Association. Flyvbjerg, B. (2006). Five misunderstandings about case-study research. Qualitative Inquiry, 12(2), 219-245. Frank, J.H. & Chandler, G. (2016). Implementing a Blueprint for District-Led School Turnaround: Cases of Three Districts. Paper presented at the annual meeting of the American Educational Research Association, Washington D.C. Frank, K. A. (1995). Identifying cohesive subgroups. Social Networks, 17(1), 27-56. Frank, K. A., Kim, C., & Belman, D. (2010). Utility theory, social networks, and teacher decision making. In A. J. Daly (Ed.) Social network theory and educational change (pp. 223-242). Cambridge, MA: Harvard Educational Press. Frank, K. A., Penuel, W. R., & Krause, A. (2015). What Is A “Good” Social Network For Policy Implementation? The Flow of Know-How for Organizational Change. Journal of Policy Analysis and Management, 34(2), 378-402. Hager, P. (2005). Current theories of workplace learning: A critical assessment. In Bascia, N., Cumming, A., Datnow, A., Leithwood, K., & Livingston, D. (Eds.) International Handbook of Educational Policy (pp. 829-846). Springer Netherlands. Hamilton, M. P., Heilig, J. V., & Pazey, B. L. (2014). A nostrum of school reform? Turning around reconstituted urban Texas high schools. Urban Education, 49(2), 182-215. Hassel, E. A., Hassel, B. C., Arkin, M. D., Kowal, J. M., & Steiner, L. M. (2006). School Restructuring Under No Child Left Behind: What Works When? A Guide for Education Leaders. Learning Point Associates. Herrenkohl, L. R. (2008). Sociocultural theory as a lens to understand organizational learning. American Journal of Education, 114(4), 673-679. 269 Higgins, M., Ishimaru, A., Holcombe, R., & Fowler, A. (2012). Examining organizational learning in schools: The role of psychological safety, experimentation, and leadership that reinforces learning. Journal of Educational Change, 13(1), 67-94. Hoppa, A. (2017, August 1). Installing the Blueprint [PowerPoint slides]. Retrieved from https://docs.google.com/presentation/d/15iIKFuKIxmc3mpUtYackTZ3P_Tv43eMWBCT i9CUTaoo/edit#slide=id.p7. Hoppa, A. (2017, August 2). How is a Blueprint District Different from a Traditional District? [PowerPoint slides]. Retrieved from https://docs.google.com/presentation/d/1Lm7hKhV3efPRQ4waU8HyeeDweywc63KdIsff 27vbrkg/edit#slide=id.p7. Irving, C. & Mohney, K. (2017, August 1). Building a System for Curriculum and Assessment [PowerPoint slides]. Retrieved from https://docs.google.com/presentation/d/1Gk6DvVaKHC4vG4lmLja- pkWN4ReCdIbxGRcox5sTsPk/edit#slide=id.p7. Klein, A. (2015). ESEA reauthorization: The Every Student Succeeds Act explained. Education Week. Knapp, M. S. (2008). How can organizational and sociocultural learning theories shed light on district instructional reform?. American Journal of Education, 114(4), 521-539. Kutash, J., Nico, E., Gorin, E., Rahmatullah, S., & Tallant, K. (2010). The school turnaround field guide. Social Impact Advisors. Retrieved November, 16, 2010. Leithwood, K., Leonard, L., & Sharratt, L. (1998). Conditions fostering organizational learning in schools. Educational Administration Quarterly, 34(2), 243-276. Lewis, C. (2015). What is improvement science? Do we need it in education?. Educational Researcher, 44(1), 54-61. Liberato, J. (2017, August 2). Instructional Infrastructure [PowerPoint slides]. Retrieved from https://docs.google.com/presentation/d/1EvlXZdq5ycy_mto- KN8KJCVn8V6m6b6z17K2oFABJUg/edit#slide=id.p7. Lin, N. (2002). Social capital: A theory of social structure and action (Vol. 19). Cambridge University Press. Louis, K. S. (2007). Changing the culture of schools: Professional community, organizational learning, and trust. Journal of School Leadership, 16, 477-487. Malen, B., Croninger, R., Muncey, D., & Redmond-Jones, D. (2002). Reconstituting schools: 270 sourcebook. SAGE Publications, Incorporated. Mohrman, S. A., Tenkasi, R. V., & Mohrman, A. M. (2003). The Role of Networks in Fundamental Organizational Change A Grounded Analysis. The Journal of Applied Behavioral Science, 39(3), 301-323. Moolenaar, N. M. (2012). A social network perspective on teacher collaboration in schools: Theory, methodology, and applications. American Journal of Education, 119(1), 7-39. Moolenaar, N. M., & Sleegers, P. J. (2010). Social networks, trust, and innovation: The role of relationships in supporting an innovative climate in Dutch schools. Social network theory and educational change (97-114). Cambridge, MA: Harvard Educational Press. Moolenaar, N. M., Sleegers, P. J., & Daly, A. J. (2012). Teaming up: Linking collaboration networks, collective efficacy, and student achievement. Teaching and Teacher Education, 28(2), 251-262. Murphy, J., & Torre, D. (2014). Creating productive cultures in schools: For students, teachers, and parents. Corwin Press. “Testing” the “theory of action”. Educational Evaluation and Policy Analysis, 24(2), 113-132. Marks, H. M., & Louis, K. S. (1999). Teacher empowerment and the capacity for organizational learning. Educational Administration Quarterly, 35(5), 707-750. McDonnell, L. M., & Elmore, R. F. (1987). Getting the job done: Alternative policy instruments. Educational Evaluation and Policy Analysis, 9(2), 133-152. McGrath, C., & Krackhardt, D. (2003). Network conditions for organizational change. The Journal of Applied Behavioral Science, 39(3), 324-336. MI Excel Statewide Field Team (2017a). Communications Driver System-Evidence of Practice [PDF document]. Retrieved from https://miexcelresourcecenter.org/wp- content/uploads/2017/12/Evidence-of-Practice-Communications-Driver-System.pdf. MI Excel Statewide Field Team. (2017b). Instructional Infrastructure-Evidence of Practice [PDF document]. Retrieved from https://miexcelresourcecenter.org/wp- content/uploads/2017/12/Evidence-of-Practice-Instructional-Infrastructure.pdf. MI Excel Statewide Field Team (2017c). Performance Management Driver System – Evidence of Practice [PDF document]. Retrieved from https://miexcelresourcecenter.org/wp- content/uploads/2017/12/Evidence-of-Practice-Performance-Management-Driver- System.pdf. Miles, M. B., Huberman, A. M., & Saldana, J. (2013). Qualitative data analysis: A methods 271 Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational advantage. Academy of Management Review, 23(2), 242-266. Neal, D., & Schanzenbach, D. W. (2010). Left behind by design: Proficiency counts and test- based accountability. The Review of Economics and Statistics, 92(2), 263-283. Nicolini, D., & Meznar, M. B. (1995). The social construction of organizational learning: conceptual and practical issues in the field. Human Relations, 48(7), 727-746. Patton, M.Q. (2002). Qualitative Research and Evaluation Methods. Thousand Oaks, CA: Sage. Penuel, W., Riel, M., Krause, A., & Frank, K. (2009). Analyzing teachers' professional interactions in a school as social capital: A social network approach. Teachers College Record, 111(1), 124-163. Resnick, L. B., & Spillane, J. P. (2006). From individual learning to organizational designs for learning. Instructional psychology: Past, present and future trends. Sixteen essays in honor of Erik De Corte, 259-276. Rice, J., Bojorquez, J.C., Diaz, M., Wendt, S. & Nakamoto, J. (2014). Evaluation of Michigan’s School Improvement Grant: Outcomes After Three Years. Retrieved from http://www.michigan.gov/documents/mde/Final_MDE_SIG_Outcome_After_Three_Yea rs_483504_7.pdf Rice, J. K., & Malen, B. (2003). The human costs of education reform: The case of school reconstitution. Educational Administration Quarterly, 39(5), 635-666. Robinson, W. S., & Buntrock, L. M. (2011). Turnaround Necessities: Basic Conditions for an Effective, Sustainable, and Scalable School Turnaround. Education Digest: Essential Readings Condensed for Quick Review, 77(3), 4-8. Rogers-Chapman, M. F. (2013). Obama's Education Plan: The Theory of Action Behind Turning Around America's Worst Schools. Education and Urban Society, 0013124513497789. Schechter, C., & Qadach, M. (2012). Toward an Organizational Model of Change in Elementary Schools The Contribution of Organizational Learning Mechanisms. Educational Administration Quarterly, 48(1), 116-153. Spillane, J. P., Hopkins, M., & Sweet, T. M. (2015). Intra-and interschool interactions about instruction: Exploring the conditions for social capital development. American Journal of Education, 122(1), 71-110. Smylie, M.A. (2016). Commentary: Three Organizational Lessons for School District Improvement. In A.J. Daly & K.S. Finnigan (Eds.), Thinking and Acting Systemically: Improving School Districts Under Pressure (pp.11-48). Washington DC: American Educational Research Association. 272 Tenkasi, R. V., & Chesmore, M. C. (2003). Social networks and planned organizational change the impact of strong network ties on effective change implementation and use. The Journal of Applied Behavioral Science, 39(3), 281-300. The Daily Cafe (2018). Literacy CAFE Menu. Retrieved from Trujillo, T. (2012, June 14). The paradoxical logic of school turnarounds: A Catch-22. Teachers https://www.thedailycafe.com/articles/literacy-cafe-menu. College Record. Retrieved from http://www.tcrecord.org ID Number: 16797 Trujillo, T. (2016). Learning From the Past to Chart New Directions in the Study of School District Effectiveness. In A.J. Daly & K.S. Finnigan (Eds.), Thinking and acting systemically: Improving school districts under pressure (pp.11-48). Washington DC: American Educational Research Association. Wohlstetter, P., Buck, B., Daly, A.J., & Smith, C.O. (2016). Common Core, Uncommon Theory of Action: CEOs in New York City Schools. In A.J. Daly & K.S. Finnigan (Eds.), Thinking and acting systemically: Improving school districts under pressure (pp.147- 182). Washington DC: American Educational Research Association. Wong, K.K.. (2016). Commentary: Toward Systemic Reform in Urban School Districts. In A.J. Daly & K.S. Finnigan (Eds.), Thinking and acting systemically: Improving school districts under pressure (pp. 221-228). Washington DC: American Educational Research Association. Yin, R. K. (2013). Case study research: Design and methods. Sage publications. 273