PRINCIPAL-AGENT RELATIONS IN OREGON EDUCATION POLICYMAKING: THE CASE OF FULL-DAY KINDERGARTEN by SUSAN RIEKE-SMITH A DISSERTATION Presented to the Department of Educational Methodology, Policy, and Leadership and the Graduate School of the University of Oregon in partial fulfillment of the requirements for the degree of Doctor of Education June 2015   ii   DISSERTATION APPROVAL PAGE Student: Susan Rieke-Smith Title: Principal-Agent Relations in Oregon Education Policymaking: The Case of Full- day Kindergarten This dissertation has been accepted and approved in partial fulfillment of the requirements for the Doctor of Education degree in the Department of Educational Methodology, Policy, and Leadership by: Dr. David Conley Chair Dr. Joanna Smith Core Member Dr. Salam Noor Core Member Dr. Lawrence Picus Core Member Dr. Brigid Flannery Institutional Representative Original approval signatures are on file with the University of Oregon Graduate School. Degree awarded June 2015   iii   Copyright 2015 Susan Rieke-Smith   iv   DISSERTATION ABSTRACT Susan Rieke-Smith Doctor of Education Department of Educational Methodology, Policy, and Leadership June 2015 Title: Principal-Agent Relations in Oregon Education Policymaking: The Case of Full- day Kindergarten The exercise of federal and/or state power is inherent to policymaking. The principal-agent theory, borrowed from economics, describes the difficulties in motivating one party (agent) to act in the best interests of the other party (principal). The theory provides insights into the roles of self-interested choice, information asymmetry, and sense making in political relationships. The extent to which the state understands the inherent challenges expressed in this dynamic and is responsive to the local school district’s specific circumstances is not well understood and thus presents an opportunity for research. This mixed methods study uses a confirmatory approach to analyze Oregon’s 40-40-20 education reform legislation and the state’s ability to operationalize education reform through the principal-agent framework, focusing on the implementation of full-day kindergarten legislation.   v   CURRICULUM VITAE NAME OF AUTHOR: Susan Rieke-Smith GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED: University of Oregon, Eugene George Fox University, Newberg, Oregon Pacific Lutheran University, Tacoma, Washington DEGREES AWARDED: Doctor of Education, 2015, University of Oregon Master of Arts, Teaching, 2000, George Fox University Bachelor of Science, Nursing, 1979, Pacific Lutheran University AREAS OF SPECIAL INTEREST: Education Policy Education Leadership Curriculum and Instruction PROFESSIONAL EXPERIENCE: Interim Superintendent, Springfield Public Schools, Springfield, Oregon 2015- Present Assistant Superintendent, Springfield Public Schools, Springfield, Oregon 2014- 2015 Director Instructional Services, Salem Keizer SD, Salem, Oregon, 2011-2014 Principal Houck Middle School, Salem Keizer SD, Salem, Oregon, 2007-2011 Assistant Principal, McKay High School, Salem Keizer SD, Salem, Oregon, 2004-2007 5th Project Coordinator, Center for Brain Injury Research and Training University of Oregon, Eugene, Oregon, 2012- Present GRANTS, AWARDS AND HONORS: NASSP Oregon Middle School Principal of the Year, 2011     vi   TABLE OF CONTENTS Chapter Page I. INTRODUCTION ..................................................................................................... 1 Agency Theory .......................................................................................................1 Cognitive Theory .................................................................................................. 3 Conceptual Policy Framework .............................................................................. 5 Fractured State Policy Systems ............................................................................. 8 Top-Down and Bottom Up Policymaking ............................................................ 9 Education Reform in Oregon .............................................................................. 10 Top-down Reform History ............................................................................ 12 40-40-20 Vision .............................................................................................13 Top-down and Bottom-up Hybrid .................................................................14 II. METHODS AND ANALYSIS ............................................................................... 18 Units of analysis .................................................................................................. 19 Time .................................................................................................................... 19 Setting ................................................................................................................. 19 Participants .......................................................................................................... 20 Data Collection Instruments ............................................................................... 20 Surveys .......................................................................................................... 20 Interviews ...................................................................................................... 22 Field Notes .................................................................................................... 23 Document Analysis ....................................................................................... 24   Data Analysis and Interpretation .........................................................................26 III. FINDINGS AND DISCUSSION ............................................................................ 28 ODE and OEIB ................................................................................................... 28 District Planning ................................................................................................. 33 Policy Agenda Communication Model ............................................................... 38 Self-Interest and Sense Making .......................................................................... 40 Theoretical Framework Critique ......................................................................... 43 Limitations and Threats to Validity .................................................................... 44 IV. IMPLICATIONS ..................................................................................................... 49 APPENDICES A. OREGON DEPARTMENT OF EDUCATION NEWSLETTER MAY 2013 .. 53 B. ALIGNMENT BETWEEN RESEARCH QUESTIONS AND DATA COLLECTION INSTRUMENTS ..................................................................... 54 C. FULL-DAY KINDERGARTEN INFORMATION SURVEY ......................... 57 D. INTERVIEW QUESTIONS: SUPERINTENDENTS ..................................... 60 E. INTERVIEW QUESTIONS: ODE STAFF ..................................................... 61 vii   Chapter Page F. SURVEY DATA ............................................................................................... 63 G. ATLAS.TI QUOTATION OUTPUT ................................................................ 79 H. RESEARCH SURVEY QUESTION MAP ...................................................... 95 I. OEIB PROPOSED SCHOOL FUNDING PRIORITIES 2015-17 ................... 96 REFERENCES CITED ................................................................................................... 97 viii   LIST OF TABLES Table Page   1. Superintendent Beliefs Regarding ODE/OEIB Supports ...................................................... 29 2. Information Sources for Guiding District Full-day Kindergarten Planning .......................... 34 3. Document and Legislative Testimony Analysis .................................................................... 35 ix   LIST OF FIGURES Figure Page 1. Triadic Structure of the Principal-Agent Model ..........................................................6 2. Triadic Structure of Oregon House Bill 3233 Principal-Agent Relation ...................16 3. Research Study Design: Qualitative Data Analysis Approach .................................27 4. Policy Agenda Communication Model ......................................................................39   1 CHAPTER I INTRODUCTION Policy is the expression of values. Fowler (2013) broadly defines policy as the dynamic and value-laden process through which a political system handles a public problem. In the interest of operationalizing values, policies work to combine governmental resources to drive implementation on the part of different public, and sometimes private, policy actors into the service of political objectives (Fowler, 2013; McDonnell & Elmore, 1987). The exercise of federal and/or state power is inherent to policymaking. Muth (1984) defined power as the ability of one actor to affect the behavior of another actor. Those who write about power relationships concentrate on the agency or power of the actors (Fuchs, 2001; Kipo, 2013; Miller, 2005; Shapiro, 2005). Coleman (1994) states the principal-agent relation emerges whenever an individual must depend on the actions of another. The exercise of power in a relationship creates potential risks for both parties. This risk-sharing problem arises when cooperating parties have different attitudes toward the potential risk (Eisenhardt, 1989; Mitnick, 1992; Perrow, 1986). Agency Theory A general property of the principal-agent relation is that one actor carries out actions which are intended to fulfill the interests of the first party, providing the first party with an extension of self (Coleman, 1994, p.146). The principal-agent problem describes the difficulties in motivating one party (agent) to act in the best interests of the other party (principal) rather than the agent’s own interest. Challenges to action arise when the two parties have differing goals and/or levels of information. These challenges 2 result in the principal’s inability to ensure the agent will act in the principal’s interest, especially when the agent perceives it is assuming a higher level of risk than the principal (Eisenhardt, 1989; Miller, 2005; Waterman & Meier, 1998). The theory has its roots in economics and thus is considered controversial by many organizational researchers (Hirsch, Michaels, & Friedman, 1987; Perrow, 1986). Perrow (1986) criticized agency theory as being ruled primarily by self-interest. He contended that it is the setting within which interactions occur that explains the behaviors of the involved parties. Some settings promote self-interested behavior, others promote other-regarding behavior and others still will be neutral. The second problem with the model was the assumption that only the agent was opportunistic. Perrow (1986) argued that unequal power exists in a hierarchical relationship. Once the concept of unequal power is entertained, opportunity for the principal to exhibit self-interested behavior exists. It is the extreme assumptions about power and preferences anchored in agency theory that limit the theory’s usefulness in describing organizational behaviors. Hirsch, et al. (1987) noted that agency theory ignored the importance of choice and the impact of ambiguity in the process of organizational decision-making. Additionally the theory failed to account for emotional responses from the agents, such as distrust, disenfranchisement, and anger, when assuming greater risks than the principal (Hirsch, et al., 1987). Eisenhardt (1989), on the other hand, asserted that much of organizational life is based on self-interest. Policy makers have the potential to cause uncontrollable variations and uncertainty in policy outcomes based on the choices they make. These choices are many times motivated by self-interest. Uncertainty introduces the inability to effectively preplan for policy implementation, thus increasing the risk for implementation on the agent’s part 3 (Eisenhardt, 1989). The theory’s usefulness, then, is in describing the conditions in which this messiness of relations can be resolved. Eisenhardt (1989) conceded that the arguments put forth in Hirsch, et al. (1987) were significant. She recommended the use of agency theory with complementary theories to capture the greater complexity of organizational behaviors. Cognitive Theory Spillane (2004) advocated for the use of theories grounded in the cognitive tradition. He argued that agency and rational choice theories assume that policy implementers get the intended policy message and thus are free to choose whether or not to implement. To be able to ignore, adapt, or fully implement depends on the degree to which the implementers understand the policy message. Understanding policy, or sense making, according to Spillane (2004), involves reconstruction of existing knowledge through interaction with the policy. Thus, implementers’ failure to do what policymakers ask them to do is more honest misunderstanding versus willful attempts to adapt policy to their own ends. Sense making depends on the information resources at hand (Spillane, 2004). Individual knowledge, expertise, and experience are several such resources. Spillane (2004) noted the level of the local policymakers’ knowledge and experience is a critical factor in the sense-making process. McPhee and Bronstein (2002) agree stating sense making is grounded in constructivism and social construction. Learning is both individual as well as collective. Social networks, such as professional or advocacy groups, provide valuable sources of information. Distribution of resources such as fiscal resources, technical support, and time for implementation are the variables that are important to consider in a cognitive model. 4 Fullan and Miles (1992) and Fullan (1996) posit implementation of any reform policy is loaded with uncertainty. In support of Spillane’s sense making notion, the early period of implementation is difficult as it requires new learning and making of meaning not just on the part of the implementer but also those initiating the reform. Fullan and Miles (1992) stress the more complex a reform, the greater the implementer’s anxiety and perceived risk. Ensuring successful implementation, however, requires conditions that support learning across the implementation continuum. Fullan (1996) describes two strategies in support of the reform system learning. The first is networking. This strategy assumes that all policy actors need integrated or coherence-making mechanisms that provide opportunities for and pressure to increase capacity and skill with the new reform. The second is reculturing or restructuring the environment within which the reform will be implemented. This requires the implementers to develop new values, beliefs and/or norms. Fullan (1996) states school cultures are adversative to change and have a vested interest in maintaining the status quo. Attempting system change through regulation and mandates inevitably results in failure as it is the people within the system that create and sustain change. McDonnell and Elmore (1987) concur citing capacity-building and systems- changing strategies as viable options for policy makers. They support Spillane and Fullan’s assertions that mandates and inducements fail as effective change levers due to lack of skill, knowledge and competence on the implementer’s part. Capacity-building is not without its own issues. By its very nature, capacity-building is often intangible and uncertain, creating anxiety for policy makers. There is an inclination on the part of policy makers to focus on the immediate utility and discount the longer term benefits. 5 While the policy actors across the continuum are the initial beneficiaries, ultimately it is the members of society that are the long-term beneficiaries of effective policy reform. Capacity-building is critical in that it offers both short and long term benefits (Fullan, 1996; Fullan & Miles, 1992; McDonnell & Elmore, 1987). McDonnell and Elmore (1987) also posit the importance of transferring official authority from policy maker to policy implementer as a function of policy implementation. A fundamental property of system-change is the distribution of authority in service to greater efficiency and effectiveness of policy reform. The alteration of authority distribution results in either narrowing or broadening participation of the various policy actors. This in turn critically alters the nature of what is produced through policy or the efficiency with which it is produced. The challenge that exists for system-changes lies in the fact that existing systems, e.g. state or districts, can blunt or co-opt attempts to distribute authority through the system. The end result is incremental modifications of existing institutions rather than whole scale reform (McDonnell and Elmore, 1987). Conceptual Policy Framework Given, then, that policy makers have the potential to introduce risk into the relationship with the implementer through behavioral choices, a conceptual framework for examining the policy actors’ behaviors in the formation and implementation of policy is needed. I propose to use an agency theory model in tandem with constructivist cognitive theory as that framework. The intent is to capture the challenges of information asymmetry and resulting risks for implementers as well as sense-making efforts made across the principal agent continuum. Mitnick (1992) criticized the use of information asymmetry and goal conflicts as constants, as policymaking concerns interactions 6 between organizations versus a simple superior-subordinate dyad. Braun (1993) agreed, stressing the importance of third-party interactions as essential to the principal-agent model. He proposed a triadic model for policy analysis. In his model, the political system (principal) obtains information about the system (third party) performance towards the principal’s goals through an intermediary agent (science-funding agencies). The third party obtains resources for its responsiveness to the principal’s goals mediated by the agent. The agent’s primacy of position in the relationship is solidified by its ability to move the third party to perform (Braun, 1993, p.140). Figure 1 illustrates an adaption of Braun’s (1993) triadic model as it pertains to this study, describing the relation between a political system, the intermediary state agents, and the local public education system. Figure 1: Triadic structure of the principal-agent model Braun, D. (1993). Who governs intermediary agencies? Principal-agent relations in research policy-making. Journal of Public Policy, 13, p. 141. Retrieved from http://www.jstor.org/stable/4007501. The challenge for the agent is whether it can build a relationship and secure the cooperation of the third party. In this model, the third party is the most powerful part in the principal-agent relation. It is in the agent’s best interest to promote the interests of the third party in order to foster the interests of the principal (Braun, 1993). Politicians are interested in re-election. They understand that optimal policy results and efficient use of Political System P Intermediary State Agent A Local School Districts T Resources for T Payment for A Service from A Performance by T Resources for T Performance by T 7 fiscal resources is directly related to their chances for re-election. The dilemma for intermediary agents is to determine the right implementation drivers to achieve the principal’s goals (Conley, 2003). Scholz (1991) and Shapiro (2005) argued that principals do not want to provide any flexibility to a third party that is exhibiting minimal compliance, while third parties will not want to be more cooperative in the face of punitive policies. Too much structure results in rigid compliance behaviors from districts and schools, thus missing the goals for reform. Too little structure results in unclear reform goals, leading educators to either wait for more information or fill the information gap with their own interpretation (Conley & Goldman, 2000). This underscores the impact of information asymmetry and need for sense making. Braun (1993) stated that parties gain information either through observation of the outcome of the other’s actions or as transmitted by the intermediary agent. Conley and Goldman (1995) noted state policymaking and implementing structures do not send coherent or consistent messages to the field, leading to skepticism on the part of districts that a policy can be successfully implemented. Spillane (2004) stressed information is gained not merely by observation of outcome but from the context within which the outcome occurs. Eisenhardt’s (1989) argument regarding the impact of policy choices by policymakers on the implementers’ ability to preplan for effective implementation has significance. Given these challenges, can state legislative action, as mediated though intermediary agencies incent public education reforms, particularly in a state with a strong tradition of local control and school site based decision-making? The extent to which the state intermediary agents understand the dynamics that exist between themselves and districts with policy implementation and are responsive to the districts’ 8 specific circumstances is not well documented in literature and presents an opportunity for research (Conley & Goldman, 1995; Dahill-Brown & Lavery, 2012; Louis, Febey, Gordon, Meath, & Thomas, 2006; Scribner, Aleman, & Maxcy, 2003). Fractured State Policy Systems The Oregon ESEA Flexibility Request (2012) noted that while pockets of public educational excellence are seen throughout Oregon, a more innovative and seamless education system that integrates state policy with local needs is needed across the board. The desire for a seamless system is not a novel one. Elmore (1996) stated that the connection between big ideas, or policy, and implementation at the instructional core is central to change in practice. He proposed that innovations requiring large system changes in the core of educational practice rarely penetrate to the classroom due to fractured systems of state and local support. The effects of this fracture are further exacerbated by legislatures and state executives engaging a limited range of policy levers as a response to perceived lack of progress relative to a particular policy, most typically mandates and inducements (Ingram & Schneider, 1990; McDonnell & Elmore, 1987). In contrast to Elmore (1996), Pogrow (1996) suggested that the fate of new reforms rests more on a common misunderstanding of the fundamentals of systems change. Reforms that become successful innovations represent a clearly defined solution to a clearly defined problem, are innovations that start small and try to accomplish one thing, and, if knowledge-based, can only succeed if all the knowledge needed is available to all. Pogrow (1996) submitted that these conditions are consistently violated by every new idea for change that is currently in play in education reform policy. Fullan (1996) proposed that as educational reforms occur within dynamically complex social structures, 9 change is inherently nonlinear and as such unfolds in a broken fashion, given all the forces that impact the change process. This appears to support both Elmore’s and Pogrow’s arguments by noting the impact of planned and unplanned policy changes emanating from fractured state systems. Top-Down and Bottom-Up Policymaking Elmore (1980) noted improved policy could be produced if policymakers would first consider whether their decisions could be feasibly implemented before they determined a course of action. Honig (2004) and Elmore (1980) framed this approach as a bottom-up or backward mapping approach. This approach requires policymakers to consider a logically ordered sequence of questions prior to policy development and choice of lever, an approach that contrasts the top-down or forward mapping approaches typically employed by federal and state policymakers. Honig (2004) asserted that the many reasons for disappointing results in educational reform policy are the product of the top-down approach typically taken by states, which does not take into account the districts’ and/or schools’ capacity for change. As previously discussed, the political cultures in which policymakers reside determine the process and levers used in policymaking. Sabatier (1986) was more cautious, stating the danger in overemphasizing the bottom-up over the top-down approach as superior in all policymaking contexts. Considering the cultures, each group is likely motivated by different concerns and thus might have developed different and more appropriate responses to policy. Wood and Waterman (1993) note that bureaucratic organizations respond to external forces, both top-down and bottom-up. According to Sabatier (1986), top-downers are more concerned with the effectiveness and efficiency of 10 a specific policy and the ability of the policy actors to guide and constrain behaviors. Bottom-uppers are less concerned with the formal enactment of the policy and more concerned with the problems the policy creates for the end user. Honig (2004) contended that either a top-down or a bottom-up approach could be mitigated in education systems when policymakers make a concerted effort to understand the capacity needs of the district and schools and use that information to provide support within the policy. Keedy and McDonald (2007) referenced the potential use of the bottom-up approach by the state when they described the SEA as a sleeping giant with untapped potential for understanding the needs and building instructional capacity in public schools. Dahill-Brown and Lavery (2012), as well as Keedy and McDonald (2007), noted a paucity of backward mapping in policy formation and, thus, in capacity building and system-changing policies at the state level. Spillane, Reiser, and Reimer (2002) noted the lack of focused sense making on the part of principals to agents and ultimately third parties, regarding the underlying principles of a given policy. Expecting third parties to adopt practices without understanding or fully constructing the underlying rationale promotes, according to Spillane et al. (2002), lethal policy mutations on the third parties’ part. Education Reform in Oregon How a state exercises influence in the policy process indicates the level of leadership the various policy actors are willing to express. Within the context of education, this means a focus on the interaction of educational stakeholders--such as parents, unions, and private and public foundations--with legislators and state regulators. 11 In the political realm, one might view the legislative and executive branches of state government as agents of the aforementioned members of the electorate. In the case of education, Oregon has an established history of local control, comprehensive policy reform, and broad grassroots participation in education policy (Keedy & McDonald, 2007; Louis et al., 2006). The progressive and open state political culture is evident in its commitment to bring stakeholders to the table for discussion as well as to provide direction for policy makers (Louis et al., 2006). Of note is the ongoing relationship between the state and stakeholders including the Oregon Education Association (OEA), the Oregon Business Council (OBC), the Oregon Business Association (OBA), Oregon Chalkboard Project, Confederated Association of School Administrators (COSA), and Oregon School Board Association (OSBA). These groups represent the most active, and therefore most visible, stakeholders engaged in providing information and recommending issues to which legislators should attend during a given legislative session. As Spillane (2004) noted, advocacy groups provide social information resources for sense making. Dependent on timing, quantity, and quality of information, these groups can either enable or limit the implementers’ sense-making ability. It is also important to note the state funding environment in which education policy is formed. In 1991 Oregon faced a fiscal dilemma, which it continues to face today, in funding education. Oregon historically funded education through local property taxation, as the state has no sales tax. Before 1990, local property taxes provided two- thirds of each school district’s funding, one of the highest rates in the nation (Goldman & Conley, 1995, p.4). In 1990, Oregon voters approved Measure 5, which consecutively 12 reduced school property tax rates over a five-year period. During that period, inequities in school district funding became apparent. The state legislature increased its role in providing state funding to level funding for school districts. State-controlled funding brought legitimacy to the state’s role in dictating programs to districts, thus diminishing the district local control. Top-down reform history In 1991, the Oregon legislature passed the Oregon Educational Act for the 21st Century (Oregon HB 3565, 1991), laying out a new vision of schooling for state public education. The legislation represented a complex “top-down” education reform intended to systematically redesign education, preschool through post-secondary education (Goldman & Conley, 1995). Oregon educators were not involved in the development of the legislation. The intent of the reform was to deliver a shock to the system rather than employ incremental change (Goldman & Conley, 1995). While the legislature saw the legislation as visionary guidance, the Oregon Department of Education saw the legislation as a blueprint for operationalization and accountability at the district level. The response from the field was almost immediate. Schools needed time and funding to plan and implement what was viewed as a major paradigm shift (Goldman & Conley, 1995). Prior to the legislation, district accountability was defined as providing a student the opportunity to learn, not ensuring learning in relation to a prescribed standard. Now districts had to determine how best to make sense of standards and determine effective instructional practices with little to no state guidance. 13 40-40-20 vision In 2011, states had the option of applying for state ESEA waivers from the US Department of Education. Past Oregon governors had not been engaged in policymaking as much as other state governors (Keedy & McDonald, 2007). This was not the case with Governor John Kitzhaber. In a renewed effort to invigorate the principles of the Oregon Education Act for the 21st Century, the governor led a coalition of legislators, educators from across K-12 and higher education, and the aforementioned stakeholders. The task was to develop a slate of system-changing and capacity-building education reform legislation to meet an aspirational college and career readiness goal for all Oregon students as called for in the Department of Education ESEA waiver application. The legislation was a major tenet of Oregon’s ESEA waiver (Oregon ESEA Flexibility Request, 2012). The legislative effort mirrored that of other states such as Georgia and North Carolina (Georgia Department of Education, 2010; North Carolina State Board of Education, 2012). In 2011, the legislature committed to college and career readiness through the passage of Oregon Senate Bill 253 (2011), known as the 40-40-20 goal. The objective of 40-40-20 is that of all high school students graduating in 2025, 40% will eventually hold a bachelor’s or advanced degree, 40% will have an associate’s degree or a meaningful postsecondary certificate, and 20% of adult Oregonians will hold only a high school diploma or equivalent and directly enter the work force. While Oregon’s post-secondary educational attainment rates have slowly improved, the passage of the goal into law prompted a new drive for collective action and reform (Oregon University System, 2011). Education and legislative leaders demonstrated strong support for the concept of a seamless, performance-based education system while expressing caution 14 about potential pitfalls. Joint Ways and Means Co-Chair Representative Betty Komp voiced enthusiasm as well as caution. “This is a political chance, it’s a fiscal chance, and it is a power chance. It’s wonderful to see but there are a lot of unknowns” (Melton, 2011). Top-down and bottom-up hybrid Unlike previous state education policy reform efforts, Senate Bill 253 and the Oregon ESEA Flexibility Request (2012) process signaled a departure from use of mandate levers to the use of systems-change and capacity-building levers through its theory of action. A theory of action framework committed the state to three overarching strategies: (a) creation of an integrated and coordinated public education system pre- kindergarten to Grade 20, (b) focused allocation of state resources on educational achievement, and (c) construction of statewide support systems to support achievement of educational goals (Oregon ESEA Flexibility Request, 2012). These strategies imply a willingness on the part of the state to entertain a blended top-down/bottom-up approach -- top-down in that the state must craft policy within federal constraints to meet the ESEA waiver requirements and bottom-up in recognition that resources and support must be invested to achieve the state’s vision. It is one thing for the state to identify strategies for systems change and capacity building and another to operationalize them. Spillane et al. (2002) noted that policies are not monolithic. The difficulty in restructuring belief systems through policy occurs when some agents and third parties view a set of policies as pressing for tremendous behavioral change while other agents and third parties view the same policies as not requiring significant behavioral change. This was certainly true with the fractured implementation of the Oregon Education Act. Spillane et al. (2002) 15 state that this response on the part of the agent and third party depends on how similar the practices required by the policy are to current practice. Between 2009 and 2013, the state legislature passed a comprehensive slate of legislation aimed at providing resources and support systems to realize the 40-40-20 goals. Four pieces of legislation are the focus of this study. First, Oregon Senate Bill 44 (2009) established a state level Full-day Kindergarten Implementation Committee in support of equitable educational outcomes for all Oregon students. The committee developed programmatic and funding recommendations to assist district implementation of full-day kindergarten programs by the 2015-16 school year. Second, Oregon Senate Bill 248 (2011) mandated half-day kindergarten and offered full weight of student funding for those districts implementing full-day kindergarten beginning fall of 2015. Third, Oregon Senate Bill 909 (2011) created an “efficient, accountable, and integrated birth-to-20 funding and governance system for public education, from early childhood services through post-secondary education and training”(Section 1.1). This bill established the Oregon Education Investment Board (OEIB) for the purpose of ensuring that all public school students in this state reach the education outcomes established for the state. Senate Bill 909 (2011) reinforced the equity of outcome policy articulated in Senate Bill 44 (2009) and Senate Bill 248 (2011). OEIB's vision is to advise and support the building, implementation, and investment in a unified public education system in Oregon that meets the diverse learning needs of the state’s youngest Oregonians through post-secondary students (OEIB, 2013). Last, in 2013, the state legislature passed legislation to establish the network of quality teaching and learning (Oregon House Bill 3233, 2013). This legislation directed 16 OEIB and the Oregon Department of Education (ODE) to establish a relationship that produces a culture of leadership and collaborative responsibility for advancing the profession of teaching among providers of early learning services, teachers and administrators in kindergarten through Grade 12, education service districts and teacher education institutions. The directive in House Bill 3233 articulates the relationship between ODE and OEIB as collaborative versus individual intermediary agents. Rob Saxton, ODE deputy superintendent at the time, echoed the collaborative nature of ODE and OEIB’s relationship as well as articulating a mission statement of service to support district implementation of state education reforms (See Appendix A). Given the various state and local policy actors included in this legislation and in consideration of Eisenhardt’s (1989) assertion that risks of implementation are potentially mitigated through preplanning on the implementer’s part, Braun’s (1993) triadic framework is useful for policy analysis of these bills. The multiple principals-multiple agents, as well as information sources, will be treated as variables rather than as constants in this model. Specifically, Spillane’s (2004) assertion regarding the impact of advocacy group information as a resource for districts must be taken into account. Figure 2 outlines this model. Figure 2: Triadic structure of Oregon House Bill 3233 principal-agent relation Political System: 1. Oregon Governor 2. Oregon Legislature P Intermediary Agents: 1. OEIB 2. ODE A Third Party: Local School Districts T Resources for T Funding to A Build Capacity Resources for T Accountability to A for use of resources Accountability to P for T performance 17 This study analyzes how and how well ODE and OEIB have operationalized the recommendations of the ODE Full-day Kindergarten Implementation Committee as intermediary agents in support of local school districts’ capacity to prepare to implement a full-day kindergarten program in their districts. My study addresses the following research questions: 1. To what extent do the OEIB and ODE work collaboratively as directed in Oregon HB 3233 and Oregon SB 909 to provide information and resources that operationalize a full-day kindergarten program? 2. To what extent do district leaders feel prepared for full-day kindergarten implementation as a result of their relationship with ODE and OEIB? 18 CHAPTER II METHODS AND ANALYSIS My interest was in understanding the behaviors and choices of the various policy actors in the preparation to operationalize full-day kindergarten within the principal- intermediary-third party model. Stake (1978) encouraged evaluators to focus on the practical concerns of the stakeholder in their immediate context rather than the more abstract concerns of the remote decision makers. He argued that by focusing on the priority issues of the practitioners, evaluators could construct rich understandings of the relationships between decision makers and stakeholder through the use of both a qualitative as well as a quantitative approach. The combination or integration of both a quantitative and qualitative approach to research is referred to as a mixed methods research design (Creswell, 2014). Creswell and Plano Clark (2011) identify three basic mixed methods designs: convergent parallel mixed methods, explanatory sequential mixed methods, and exploratory sequential mixed methods. I used the explanatory sequential mixed methods approach for this study in which quantitative data was collected and analyzed first and then qualitative data was collected and analyzed. Both sets of data were then used to develop an interpretation. The underlying assumption of this approach is that quantitative and qualitative data provide different types of information, thus allowing for contradictions or incongruent findings to be explained or probed further (Creswell, 2014) as well as providing a framework for the construction of the rich relationships between decision makers and implementers recommended by Stake (1978). 19 Units of Analysis In this study, there were four units of analysis. The first was the legislative and executive branches as they represent the principals who articulated the goals for education reform. The second was the intermediary agents in Oregon state education policies and their ability to provide information that built capacity in third party organizations to effect a system change. These were the designees for the state policy chiefs for the Oregon Department of Education and the Oregon Education Investment Board. The third were the state public school superintendents as third party entities. The fourth were the social artifacts, as represented by legislative, state agency, and district documents. Time For this study, data was collected over a three-month period, January to March 2015. The danger, as Babbie (2010) points out, is group attrition over time, which has the potential to distort the results of the study. While ODE and district staff did not change during the course of the study, access to OEIB staff was restricted due to reassignments as a result of former Governor Kitzhaber’s resignation in February of 2015. The impact on the data collected is accounted for relative to the final conclusions. Setting The triadic principal-agent-third party model required observations of the policy actors in their environments. This study included data collected from multiple sites: (a) Oregon Legislature, (b) Oregon Department of Education, (c) Oregon Education Investment Board meetings, and (d) the Springfield, Bethel, and Fern Ridge school districts. 20 Participants Participants were chosen based on their specific roles within the principal-agent- third party model. They were: (a) the principals as represented by Oregon state legislative sponsors of full-day kindergarten legislation (n=1), (b) intermediary agents as represented by ODE and OEIB Early Learning K-3 staff (n=2), and (c) third parties represented by member superintendents of the COSA Policy Vision workgroup including the superintendents of the Bethel and Springfield public schools as well as the other superintendents from Lane County school districts (n=15). These four Lane County districts represent a range in district size from 3,000 to 20,000 students, which provided data regarding the potential impact of district size on the superintendent’s perception of preparedness for implementation. Data Collection Instruments In a review of studies on state policymakers (Keedy & McDonald, 2007; Louis et al., 2006; Louis et al., 2008), many of the studies used the same data collection instruments I used in this study. Guided by this body of research, I used a two-phase approach. In phase one, a mixed Likert scale forced-choice and open-ended question survey was given to both COSA Full- day Kindergarten Work Group superintendents as well as Lane County public school superintendents. Phase two data was collected through interviews guided by the survey data, data collected from the field, and document analysis. Surveys Surveys elicit information directly from study participants (Fink, 2013). I used a mixed forced-choice and open-ended survey. I considered several options in the design 21 of the survey instrument. The first was the forced-choice survey. The advantages of the forced-choice survey were ease of coding and disaggregation of data, making reporting of data straightforward. The disadvantages included the risk of influencing responses due to forced-choice construct, the order of options affecting the results, and the possibility that some response options are not always informative. The second option was the open- ended survey. The primary advantage of this survey type was its allowance for rich and in-depth narrative from the subject. The disadvantages were the time requirements for survey completion, as well as the length of time for analysis of data; the increased complexity in coding data; and potential threats to validity, as interpretations may need to be made when analyzing data. The third option was a mixed forced-choice and open- ended survey. The advantage here was the opportunity to gather data that was both easily analyzed as well as narrative from subjects regarding why the subjects responded as they did. As such, I developed a mixed forced-choice and open-ended survey to elicit data about how and why the various policy actors and implementers behaved as they did (see Appendix C). As districts are currently actively engaged in planning full-day kindergarten implementation, the survey gathered information about the subjects’ knowledge in the following areas and the impact of the information on the district planning process: (a) knowledge of Oregon Senate Bill 44 (2009) and the Full-day Kindergarten Committee recommendations, (b) beliefs about the laws’ intent, (c) beliefs about the laws’ potential effects, and (d) beliefs about the district’s capacity for successful implementation based on information received from ODE and/or OEIB. Survey data was gathered during phase one of the research from a self-administered survey distributed online to study participants. The survey was sent to the sixteen Lane 22 County school district superintendents and nine of the eleven superintendents who were members of the COSA work group. The other two members were Lane County superintendents. Fifteen superintendents responded to the survey -- eleven who were not members of the COSA workgroup and four who were members of the COSA workgroup. Thirty-one percent reported one to three years of experience as a superintendent, 15% reported three to five years’ experience, 23% reported five to ten years’ experience, and 31% reported more than ten years’ experience. Data was segmented by superintendent membership in the COSA work group. The superintendents who were not COSA work group members represented districts with student populations of less than 1,000. Superintendents who were members of the COSA workgroup represented districts from 5,000 to 10,000 students. Interviews In phase two, I conducted semistructured interviews with individual subjects (n= 6). Interview questions were developed following an analysis of the data collected from the superintendents’ survey. Questions focused on areas where superintendents perceived strengths and weaknesses in developing district implementation plans based on resources or information received from OEIB and ODE. Questions were also differentiated based on COSA Full-day Kindergarten Work Group membership. Interviews were recorded by two methods: (a) handwritten notes and (b) audiotaping. The methods used were dictated by the permission received from the Lane County school superintendents and ODE staff. Based on the superintendent survey data, six questions were developed for superintendents and ODE early learning staff (see Appendices D and E). 23 I interviewed Oregon Department of Education sponsors of Oregon Senate Bill 44 (2009) and Oregon Senate Bill 248 (2011) and I interviewed the superintendents from the Bethel, Fern Ridge, and Springfield school districts. Superintendents from Bethel and Springfield also served on the COSA Early Learning and Full-day Kindergarten Policy Work Group. The Eugene superintendent’s office declined my request for interview. COSA staff also declined my request for interview, citing calendar conflicts due to the current Oregon legislative session. I received comments from the superintendents of three other Lane County school districts regarding ODE and OEIB support for full-day kindergarten. These superintendents declined to participate in the survey as they had implemented full-day kindergarten between 2006 and 2008, but shared their opinions through personal communications concerning ODE and OEIB as collaborative partners in supporting districts for this study. The superintendents represented school districts that have student populations between 250 and 500 students. I also received personal communication from Lane County legislators regarding full-day kindergarten funding. Data from school superintendents who served on the COSA work group were segmented from non-work group members. Segmentation provided a comparative dimension between the beliefs of the larger work group of superintendents and the non-workgroup superintendents. Field notes I gathered field observational data regarding preparation to implement full-day kindergarten from the Oregon School Boards Association State Fall Conference in November of 2014 and the COSA Winter Central Office Administrators Conference “Off the Record” meetings for superintendents in January of 2015. As SB 44 (2009) and SB 24 248 (2011) were enacted legislation, I reviewed and transcribed archival audio tapes (n=12) of Oregon Senate Education Committee and House Education Committee hearings on SB 44 (2009) and SB 248 (2011). I reviewed transcribed archival video of full-day kindergarten testimony given to the OEIB Best Practices Committee meetings by the Chair of the COSA Early Learning and Full-day Kindergarten Work Group in 2014. Archival records of the COSA Early Learning and Full-day Kindergarten Work Group were not available for review and analysis. Observations, as recorded through field notes, provided detailed descriptions of the setting, participants, agenda items, and decisions approved or not approved, as well as non-agenda items that arose during the course of the various meetings (Babbie, 2010; Creswell, 2014; Mulhall, 2002). Creswell (2014) noted that advantages of field observation include recording information as it occurs, the ability to notice unusual aspects during the observation, and the ability to explore topics that may be uncomfortable for participants to discuss. Mulhall (2002) expanded on this last advantage stating that observations allow for insight into interactions between dyads and among groups; in short, do they walk their talk or not? The disadvantage to gathering data from archival audio and video materials was my inability to observe in situ interactions between the committee and those providing testimony and among committee members. Document analysis Bowen (2009) recommended inclusion of document analysis as a means of data triangulation. By examining information collected through various sources, findings can be corroborated across data sets to reduce the impact of potential biases that can exist in a single data source. There are both advantages and disadvantages to document analysis 25 (Bowen, 2009). Documents provide data on the context within which the research participants operate and suggest some questions that need to be asked. Additionally, documents provide a means for tracking change over time. Subtle changes in a draft can reflect substantive developments in a policy or implementation effort (Bowen, 2009, p. 30). This ability to track change over time was useful in evaluating the triadic principal- agent model as it applies to Oregon education reform policymaking. Disadvantages to the use of document analysis include: (a) insufficient detail for research, as they were not produced for that purpose; (b) low retrievability; and (c) the opportunity for incomplete collection of documents or biased selectivity. Bowen (2009) stated these three concerns should be viewed as flaws rather than major disadvantages. The following documents (n = 30) were analyzed: (a) Oregon SB 44 (2009), Oregon SB 248 (2011), Oregon SB 253 (2011), Oregon SB 909 (2011), and Oregon HB 3233 (2013); (b) COSA Early Learning and Full-day Kindergarten Superintendents’ Workgroup final report (2014); (c) ODE Full-day Kindergarten Implementation Report (2010); ODE Superintendent Pipeline newsletters from 2009 to 2012; Deputy Superintendent Education Update newsletter from 2012 to 2015; Early Learning staff communiqués and documents concerning full-day kindergarten; (d) OEIB newsletters from 2012 to 2015; OEIB policy and budget recommendation documents; and (e) school district level full-day kindergarten team planning documents. These instruments captured state policy issues, policy actors and their influence on state policy issues, political culture and its impact on the collaboration among policy actors, and the state role in capacity building. The documents served as a frame to address the study research questions as outlined in Appendix B. 26 Data Analysis and Interpretation Data analysis in an explanatory sequential mixed methods approach included the challenge of how best to interpret the data. Creswell and Plano Clark (2011) recommended analyzing the two data collections separately and then merging the data. This was done through a side-by-side comparison. Descriptive statistical data from the forced-choice survey questions was collected first (See Appendix F). This data was compared to the qualitative findings from the open-ended survey questions, interviews, and field observations to confirm or disconfirm the descriptive statistical results. A portion of the qualitative output is shown in Appendix G. The qualitative data was analyzed using ATLAS-ti software. The software supported the coding of documents, interview transcripts, and survey data, which resulted in a concept network map for analysis. The map was compared to the descriptive statistical data to confirm or disconfirm the principal-agent model. This process is outlined in Figure 3. The bidirectional arrows between the data collection and data analysis boxes indicate the back and forth analytical process employed as the behaviors of the multiple principals and multiple intermediary agents were considered in this model. 27 Figure 3: Research study design: Qualitative data analysis approach RQ 1:To what extent does the OEIB and ODE work collaboratively as directed in Oregon HB 3233 and Oregon SB 909 to provide information and resources that operationalize a full-day kindergarten program?? RQ 2:To what extent do district leaders feel prepared for full-day kindergarten implementation as a result of their relationship with ODE and OEIB? Field Observations: OEIB Meetings COSA Superintendent Meetings Mixed forced- choice and open- ended survey: Superintendents Document Analysis: Legislation ODE Full-Day Kindergarten documents OEIB Early Learning newsletters District full-day kindergarten planning Documents Interviews: OEIB/ODE Early Learning staff Superintendents State legislators Data Analysis: Generate and analyze descriptive statistical data; Coding qualitative data: data themes; generation of concept network (ATLAS-ti software) Merge quantitative and qualitative data Final analysis and interpretation Triadic Principal-Intermediary Agent-Third Party Theory 28 CHAPTER III FINDINGS AND DISCUSSION The two questions guiding my study were: (a) To what extent does the OEIB and ODE work collaboratively as directed in Oregon HB 3233 and Oregon SB 909 to provide information and resources that operationalize a full-day kindergarten program; and (b) To what extent do district leaders feel prepared for full-day kindergarten implementation as a result of their relationship with ODE and OEIB? Results from the study indicate that superintendents in Lane County felt able to plan and guide their districts to implementation of full-day kindergarten in the fall of 2015. Superintendents reported that the ability to plan and guide the implementation was a direct result of their relation with COSA and not the result of ODE and OEIB’s collaborative efforts. The research survey question map provided a framework for detailed analysis of the survey data based on the research question. Furthermore, the map guided data comparisons between the quantitative and qualitative data sets (See Appendix H). ODE and OEIB Collaboration In answer to the first research question, the survey elicited data regarding superintendent beliefs about the ability of the ODE/OEIB to collaborate and provide technical support to districts as they planned for full-day kindergarten. Belief data was categorized by work group membership. Superintendent beliefs are reported in Table 1. 29 Table 1 Superintendent Beliefs Regarding ODE/OEIB Supports Belief Agree Disagree Split Agree/Disagree ODE/OEIB information consistent regardless of agency Non-work group Work group Superintendent knows who to contact for information to support district planning Work group Non-work group ODE/OEIB provide necessary resources for implementation Non-work group Work group ODE/OEIB adjust information and resources based on district feedback Work group Non-work group ODE/OEIB adjust information and resources based on COSA feedback Work group Non-work group ODE/OEIB adjust information based on community group feedback Work group Non-work group My district has the staff and facilities to implement full-day kindergarten Work group Non-work group Note. Non-work group Superintendents (n=11); Work group Superintendents (n=4) COSA work group members reported that ODE and OEIB adjusted information and resources to districts based on feedback from districts as well as COSA and other community groups. Work group members knew whom to contact at ODE or OEIB for additional information if needed. The group was split on whether ODE and OEIB were consistent in their messaging regardless of the agency. Interview data with work group members suggested two possible explanations for the split. First, members had a close working relationship with both ODE Deputy Superintendent Saxton and OEIB CEO Golden as they attended many, if not all, of the Full-day Kindergarten Work Group sessions. As both agency heads were former superintendents, there was a belief among the work group members that “they are one of us,” thus the messages were aligned. 30 Second, all superintendents articulated an understanding of OEIB’s mission to drive alignment of public education functions through strategic allocation of state funding. But, they also pointed to OEIB’s inability to deliver an aligned service and funding system while holding K-12 districts harmless. This may have resulted in the belief that messages were not aligned. A Lane County superintendent on the work group noted that he expected more from ODE and OEIB, especially since the deputy superintendent early in his tenure made a commitment to the COSA superintendents at Off the Record sessions and to districts across the state to be a service organization for schools. Another superintendent reported simply bypassing ODE specialists and talking directly with Deputy Superintendent Saxton when encountering inaccurate information because of the relationship she had developed during the work group experience. She further noted frustration when she had to call the deputy superintendent instead of having the ability to get “…a straight and timely answer from those specialists.” Another Lane County superintendent echoed the same frustration, faulting both ODE and OEIB for taking on too many initiatives without building adequate staff or funding to support successful planning at the district level. Non-work group members were more likely to be split about their beliefs regarding ODE and OEIB collaboration and support to districts. They disagreed that ODE or OEIB provided consistent information or support. They disagreed that either agency listened to feedback or adjusted information and resources based on feedback. Interview data did not support a personal relationship with either Deputy Superintendent Saxton or CEO Golden to the same extent as the work group superintendents. They did express the same belief as the work group superintendents that ODE and OEIB had made 31 a commitment to superintendents to be service organizations, a break from past practice. They cited frustrations regarding the number of unfunded initiatives turned mandates since the passage of the 40-40-20 legislation. Unending and unfunded mandates from the legislature were not perceived as serving districts’ best interests. They specifically referenced ODE and OEIB’s shift in position from their initial support early in the work group process for additional funding for full-day kindergarten. The work group superintendents reported that one time additional funding needs were clearly outlined in the work group report given to both OEIB and ODE in addition to the shift to full funding weight for kindergarten students. In a review of the archival video of OEIB, McMinnville Superintendent Russell, chair of the Full-day Kindergarten Work Group, provided testimony and copies of the COSA Work Group report to OEIB. In her testimony, she clearly outlined that the capital and fiscal resources needed to support full-day kindergarten implementation in districts statewide required more support than full funding weight. This was in stark contrast to testimony provided to the Oregon Senate Education Committee regarding SB 248. A member of the ODE Full-day Kindergarten Implementation Committee testified that they were unable to quantify projected capital and staffing costs for full-day kindergarten implementation. It should be noted, per one work group superintendent, that ODE budget staff provided support to the COSA work group to ascertain implementation costs. In the video, CEO Golden stated that while full-day kindergarten was an important path to third-grade reading, state resources would need to be reallocated to support early literacy pre-K to third grade. When asked by Superintendent Russell what current district initiatives should no longer receive funding to support early literacy, 32 CEO Golden replied, “That’s a local district decision.” OEIB’s proposed Early Learning budget for the 2015-17 biennium did not include an additional allocation, instead rolling up full-day kindergarten as part of the third-grade literacy initiative (see Appendix I). One superintendent stated, “No one is arguing against third grade reading or the part that full-day kindergarten plays. We’re just tired of taking all the risks for making their grand visions a reality.” Interview data from ODE Early Learning specialists presented a different perception. It should be noted there are only two ODE Early Learning specialists to support 197 school districts in Oregon. Both specialists reported a close working relationship with OEIB policy staff and repeated throughout the interviews that early learning policy, messaging, and support to districts about early literacy were aligned between the two agencies. When asked for examples, both specialists noted the administration of the district full-day kindergarten facilities needs survey. They pointed to posting the COSA Full-day Kindergarten Work Group report and curriculum resources provided by districts who were already offering full-day kindergarten, as well as policy research reports in support of full-day kindergarten on the ODE website. They cited supporting COSA full-day kindergarten conferences offered across the state. When asked to describe the relationship between ODE and COSA, both specialists described COSA as a “partner in the work” and offered no further details. A work group superintendent noted that the ODE survey was generated by the work group and offered to ODE. He continued that COSA staff actively lobbied ODE until they administered the survey to districts. 33 One specialist stated the primary focus of ODE and OEIB was to drive equity of outcomes for all students. When asked to elaborate, the specialist offered that the mission was to help districts reallocate resources to “high leverage” practices that lead to the outcomes identified by OEIB for third grade reading. Field data from the Fall OSBA 2014 conference recorded nearly identical keynote addresses, one each from Deputy Superintendent Saxton and CEO Golden on two successive days in support of the state’s Early Learning and Literacy initiatives. Additionally, breakout sessions provided by ODE Early Learning specialists on the literacy initiatives reinforced early literacy policy themes: (a) equity of outcome as measured by third grade reading proficiency rates and (b) reallocation of district funding in support of early literacy. ODE and OEIB documents gathered from the breakout sessions were similar in their policy stance, rationale for early literacy, and focus to reallocation of district resources. The OSBA conference is key to messaging and motivating education reforms, as superintendents and their boards from across the state are in attendance. In short, it is the state’s opportunity to “make sense” to districts regarding policy rationale and use of levers, in this case funding, to drive implementation. District Planning As for the second research question, the survey elicited data concerning the types and sources of information superintendents found most useful in planning. Information sources are reported in Table 2. Legislative (n=4), ODE/OEIB (n=12), COSA (n=10), and district planning documents (n=4), as well as legislative testimony (n=12) and OEIB testimony (n=1) were analyzed to track the flow of information from legislation to district plans. Document and testimony data are reported in Table 3. Themes for document and 34 testimony analysis were chosen based on the types of information typically required to successfully plan for this type of policy implementation at the district level. A count of one was given for each theme that was present and developed in each document. A count of one was given for each person providing testimony or legislators discussing the specific theme. Table 2 Information Sources for Guiding District Full-day Kindergarten Planning Primacy of Usefulness Non-Work Group Superintendents Work Group Superintendents 1 COSA Full-day Kindergarten Conferences COSA Work Group Support Staff 2 COSA Work Group Report SB 44 3 Colleague Superintendents Full-day Kindergarten Conferences 4 ODE Newsletters Colleague Superintendents 5 ODE Early Learning Specialists ODE Deputy Superintendent Saxton 6 SB 44 OEIB Support Staff 7 ODE Deputy Superintendent Saxton OEIB CEO Golden 8 OEIB Newsletters ODE Newsletters 9 OEIB CEO Golden ODE Early Learning Specialists 10 State Senator OEIB Newsletters 11 State Representative State Representative 12 OEIB Support Staff State Senator Note. Non-Work Group Superintendents (n=11); Work Group Superintendents (n=4) 35 Table 3 Document and Legislative Testimony Analysis Information Themes Legislation Legislative Testimony ODE Documents OEIB Documents COSA Documents District Planning Document Full-day Kindergarten Rationale Equity SB 44 SB 248 10 (SB 44) 4 (SB 248) 4 2 4 4 of 4 districts Full-day Kindergarten Rationale PreK-3rd grade SB 248 12 (SB 248) 3 9 4 1 of 4 districts Funding source State SB 44 SB 248 2 (SB 44) 4 (SB 248) 0^ 0^ 10^ 4 of 4 districts Funding source District N/A 6 (SB 44) 10 (SB 248) 1† 1† 0 0 of 4 districts Required Capital Resources SB 44 3 (SB 44) 4 (SB 248) 1 0 10 4 of 4 districts Staffing N/A 0 (SB 44) 4 (SB 248) 0 0 9 4 of 4 districts Implementation Timeline SB 44 SB 248 10 (SB 44) 15 (SB 248) 0 0 4 4 of 4 districts Instructional Practices N/A 10 (SB 44) 2 (SB 248) 3 (1 of 3 *) 0 1 4 of 4 districts Instructional Schedule N/A 2 (SB 44) 10 (SB 248) 1* 0 1 4 of 4 districts Curriculum Guidance N/A 0 (SB 44) 2 (SB 248) 1* 0 4 4 of 4 districts *COSA documents posted on ODE Early Learning website ^ One time additional allocation † Biennial State School Fund allocation Survey data indicated that both work group as well as non-work group superintendents found information from COSA and their colleagues more helpful than information from ODE, OEIB or other sources. The document analysis presented an interesting segmentation of information into two categories: (a) information that is critical to successful planning and (b) information that communicates policy values. 36 Critical information such as funding, capital needs, implementation timelines, instructional schedules, and curriculum guidance information sources were nearly absent in ODE and OEIB documents versus COSA documents. Value-oriented information, e.g. rationale for full-day kindergarten and equity of outcome, was present across the legislative, state, COSA, and district planning documents. Communication regarding rationale for early literacy is interesting in the fact that information was consistent across legislative, state, and COSA documents. Only one district plan included a model for reallocation of district resources for a preschool expansion as part of the full-day kindergarten plan. This may be the result of concerns regarding ability to fund full-day kindergarten, let alone a prekindergarten expansion. It may also be more the result of this specific superintendent’s previous experience from another state where they had developed and implemented a preschool to Grade three literacy program than an outcome of state guidance. In interviews with superintendents, they expressed frustration with ODE and OEIB for not being responsive to district fiscal and capital needs. As previously noted, superintendents held the expectation based on interactions with the agency heads that both OEIB and ODE were focused on serving district needs as they planned for implementation of strategic initiatives. It is important to note that the legislature directed ODE to assess funding and capital needs for district implementation of full-day kindergarten (Oregon SB 44, 2009). In a review of Senate Education Committee audiotapes, Senator Haas, sponsor of full-day kindergarten legislation, was clear that adequate resources must be provided to ensure implementation of full-day kindergarten. SB 44 directed ODE to develop funding recommendations for full-day kindergarten 37 implementation and report back to the legislature by the next biennium. ODE’s report outlined the funding mechanisms and timelines for increasing kindergarten funding to full weight for full-day students. However, signaling a change of course, during hearings on SB 248 authorizing full-day kindergarten, small district superintendents who were members of the ODE committee provided testimony in support of full-day kindergarten, regardless of access to state funding. In response to the testimony, Senator Haas shifted his support from mandating state funding to “…continuing the conversation at a later date.” As the state was in the depth of a recession, his concern was to “…meet the needs of all Oregon students” by setting a clear path for districts to “…do the right thing as they always do.” He knew mandated funding would be a barrier to passage of the legislation he was firmly committed to passing. As committee chair, Haas led a procedural vote that resulted in the bill being referred directly for a vote in both chambers, bypassing the Ways and Means committee. Two members of the committee stated their concerns for adding yet another fiscal burden on districts already in the throes of budget cuts. Members of the committee committed to revisiting the funding issue in future sessions as the state budget allowed. With the creation of OEIB later in the 2011 session and under its direction, ODE Early Learning specialists focused districts on strategies for reallocation of state funding in support of early literacy. One specialist reported facilitating many reallocation funding discussions with districts to fund full-day kindergarten in transition to early literacy goals. To frame the discussion of the findings, I return to the proposed blended agency and constructivist cognitive model. The model presumed that policy information flowed in a circular, transactional manner from the legislature to ODE and OEIB and 38 then to the district. The model also presumed information was contextualized for districts in a manner that supported planning for full-day kindergarten. The findings, however, indicate a different model. Policy Agenda Communication Model The initial triadic model for information and resource transmission implied a straightforward transactional relationship among the legislature, ODE and OEIB, and the districts. It is clear from the data that this is not the case. Transmission of information themes presented in Figure 4 may be considered in two ways, linearly from legislation to the district plan and in clusters dependent on a policy actor’s agenda. Value-laden information, e.g. equity and Pre-K-third grade literacy appeared to move in a linear fashion from legislation through ODE and OEIB communications to district plans. However, these values were more pronounced within legislative, ODE, and OEIB documents than COSA documents, thus clustering in the policymaking bodies and reflecting their policy agenda. Information critical to planning a successful implementation was most evident in COSA and district plans and sparse to nonexistent in policymaker documents, clustered in the implementer group and reflective of that agenda. 39 Figure 4 Policy agenda communication model As depicted in the model, information related to policy goals was aligned between the policymaking bodies and accurately transmitted out to districts. Based on SB 248 (2011) legislative testimony, legislators delegated policy implementation to ODE without a directive to ODE for future reports on district implementation efforts. Senator Haas’s and the Senate Education Committee’s vision for full-day kindergarten was in place and the work of the legislature done. Despite the fact that both ODE and OEIB agency heads were former superintendents with extensive experience in planning policy implementations in their former districts, both defaulted to policymaker behaviors. One Early Learning specialist noted that the longer they were away from their district, the more likely they were to allow policymaking versus district’s beliefs and dispositions to guide their work. Districts, on the other hand, looked to and expected support from these bodies to support their planning process. When districts determined direct communication with ODE and OEIB was ineffective, they turned to COSA to advocate for their needs. Legislative testimony and interview data with ODE staff confirmed COSA communications to the legislature and ODE on the districts’ behalf. There was no Legislature OEIB ODE COSA Districts Funding Capital Needs Funding Capital Needs Policy Goals Funding Capital Needs Technical Assistance Policy Goals 40 evidence the legislature nor ODE engaged COSA as a sense maker for districts. Self-Interest and Sense Making Sears and Funk (1991) suggested policy actors’ actions stem from egoistic concerns and are a form of interest-maximizing behaviors. These behaviors become even more potent when linked to one’s group’s interest where the costs and benefits of the group also affect the self. Self-oriented group interest (Sears & Funk, 1991) causes the individual group members to support a policy goal or given action because it benefits the group and by extension, the individual member. It is salient to the discussion, then, to consider the potential self-oriented interests of the groups represented in the data. Superintendents assume significant amounts of risk and responsibility for sense making in their relations with board members and line staff when planning and implementing state policy. They know from experience that each policy has potential unintended consequences that compound with each new policy. They are the sense makers for their boards, who believe local control still exists and are easily frustrated by the constraints of state mandates. A Lane County superintendent noted their board’s perception that local control ceased to exist once school funding became a state responsibility. With state fiscal control came state accountability as exercised typically through unfunded legislative mandates. They cited the impact of open enrollment legislation on a district’s decision to offer full-day kindergarten as the most recent example. All districts in Lane County will offer full-day kindergarten rather than potentially lose students, and revenue, to neighboring districts. They noted, “Legislators give lip service to local control, but every piece of new [education] legislation further restricts independent board and superintendent decision making.” Superintendents are 41 the sense makers for line staff who typically display angst when asked to implement abstract policy concepts into concrete action. Staff know from experience policy implementation requires additional fiscal resources for professional development, planning, staff, and other capital needs. These concerns were evident in all four district plans. Many districts in Lane County have gone or are in the process of going to their communities for bonding authority to cover facility and furniture costs. Unlike many other states, Oregon does not provide additional fiscal resources for school facility improvements. Despite the legislature’s contention that full-day kindergarten expansion was included in the state school fund budget for 2015-17, superintendents in Lane County indicated they would be reallocating funding from Grades 1 through 12 to provide adequate support for full-day kindergarten implementation. In the four Lane County districts studied, three will need to increase class size by 0.5 to 1 student per teacher FTE to fund kindergarten. The fourth district has local options tax revenue to offset state school fund shortfalls. Policymakers also accept risk. Given the length of his legislative career, former Governor Kitzhaber understood the potential that existed in effective reform: equity of outcomes for all students while containing costs. His desire to create and sustain an efficient and aligned education system is considered visionary by many policymakers (Melton, 2011,). One may question the wisdom of initiating many of the 40-40-20 initiatives, but not the leadership shown while guiding the legislature and state agencies towards achievement of the vision (Oregonian Editorial Board, 2015). Policymakers’ knowledge and experience is a critical factor in the sense-making process for the implementer as policy making occurs within a political context. Policymakers understand 42 and leverage power through like-minded supporters (Fowler, 2013; Sears & Funk, 1991). Former Governor Kitzhaber was purposeful in recruiting Deputy Superintendent Saxton and CEO Golden to their posts. Both had established track records in Oregon of moving districts towards equity of educational outcomes for students. In turn, Saxton and Golden purposefully recruited staffs with extensive knowledge and expertise from their former districts. The staffs were in a position to assist districts in the conversion of abstract policy concepts into concrete action steps. It is significant to note, however, the number of support staff for both agencies. With only two specialists to support early learning and full-day kindergarten, implementation across 197 districts is a daunting, if not impossible, task. Given the challenge, there was an opportunity for ODE to collaboratively engage advocacy groups such as COSA and OSBA as policy sense makers and capacity builders for superintendents and boards, thereby reinforcing their effectiveness to districts. Staff, by their own admission, were co-opted over time by those with long agency and/or policy- making history at the state and thus became entrenched in the “us versus them” mentality. To that point, advocacy groups, such as COSA, are a valuable source of information, as they understand both professional cultures where their members operate as well as the political culture of policymaking (Spillane, 2004; McDonnell & Elmore, 1987; McPhee and Bronstein, 2002). In this study, COSA presented more as a persistent, and not always welcomed, sense maker for ODE and the legislature regarding district needs. COSA’s existence is dependent upon how they are perceived in their role as advocate for district administrators and restrainer of perceived unbridled use of state authority. COSA provided systematic support structures for superintendents to both gather information for 43 planning as well as build capacity across districts through COSA full-day kindergarten conferences. COSA was purposeful in not only identifying experts in the field, but also engaging them as trainers for districts. While ODE and OEIB staff participated in the conferences, only documents from practitioner-led breakout sessions were found in Springfield, Bethel, and Fern Ridge school district planning documents. The strength of COSA’s advocacy for districts provides a cautionary tale for state policy makers. Superintendents, through COSA, are currently lobbying the legislature for the sun setting of OEIB prior to 2016. Deputy Superintendent Saxton recently announced his resignation as deputy superintendent after a two-year tenure. While the factors influencing his decision are not known, it is suggestive that superintendent dissatisfaction may be a factor. Superintendents interviewed for this study cited perceived ODE and OEIB missteps with the manner in which they supported districts through the myriad of 40-40-20 strategic initiatives, including full-day kindergarten. Theoretical Framework Critique A review of the usefulness of the theoretical framework for this study is in order. As previously stated, I proposed the use of agency theory, specifically Braun’s (1993) triadic model, in tandem with a constructivist cognitive theory as proposed by Spillane (2002). Two criteria of the triadic model must be addressed: (a) presumption of resources flowing from principal to third party and (b) presumption that the principal understands the need to secure cooperation of the third party. The Oregon legislature did not secure nor provide additional fiscal resources, either for ODE to build sufficient staff to support districts or for districts to implement full-day kindergarten without impacting current instructional programs, thus hobbling ODE and OEIB’s ability to secure 44 cooperation of districts from the outset. One may well view the triadic model as ineffective for Oregon policy making due to an established history of unfunded education mandates. Conversely, one may also view the triadic model as useful when considering Eisenhardt’s (1989) assertion that policymakers have the potential to cause uncontrollable variations and uncertainty in policy outcomes based on the choices they make. The outcome of the legislature’s choice to not delay full-day kindergarten legislation by tying it to the state budget ultimately resulted in challenges for both ODE and OEIB in their relations with districts. The use of a constructivist cognitive theory framework proved useful in guiding the analysis of the qualitative data. To this point, it proved useful in capturing the greater complexities of the various policy actor behaviors and thus a greater understanding of the relational dynamics among districts, ODE, OEIB, and COSA (Eisenhardt, 1989). The combination of an agency model with the constructivist cognitive theory should be considered as useful in policy analysis. Perrow (1986) and Hirsch, et al. (1987) were correct in their assertions that the impact of choices and emotional responses could not be captured in the agency model. The blended model provides not only a framework for understanding the impact of choices across the principal–intermediary agent-third party continuum but guidance concerning corrective actions that will improve policy implementation and outcomes. Limitations and Threats to Validity There are two significant limitations to my study: time and small numbers of participants. The study is strengthened by use of a mixed methods approach, which provided a framework for understanding the data through the construction of the rich 45 relationships between decision makers and implementers. It should be noted that these two factors significantly impacted the generalizability of this study beyond Lane County. The possibility should be considered that data gathered from Portland Metro area superintendents, where districts serve larger student populations and have greater central office supports for policy implementation, might yield different responses to the survey and interview questions. Conversely, data gathered from eastern or coastal Oregon communities may well demonstrate a greater district reliance and thus a more positive view of ODE and OEIB due to lack of central office support staff. Validity in an explanatory sequential mixed methods approach is based on establishing quantitative as well as qualitative validity for each database. To this end, it is critical that data collection used the same variables, constructs, and/or concepts (Creswell, 2014). Creswell (2014) identified several potential threats to validity in a mixed methods approach: (a) unequal sample sizes that provide less of a picture via the qualitative data than the quantitative data, (b) use of different concepts or variables on both sides making data difficult to merge, and (c) failure to account for data discrepancies when determining final conclusions. Creswell and Plano Clark (2011) recommended the following strategies be used for internal validity control: 1. Triangulation of data – Data triangulation was achieved using different sources of information in order to increase the validity of the study. Validity of the qualitative data was built upon examination of evidence from various sources and using the various forms of data to develop a coherent description and justification for the study’s themes (Creswell, 2014; Golafshani, 2003). The data sources for this study were collected from surveys, archival audio recordings of Oregon Senate and House Education 46 committee hearings, and work sessions on SB 44 (2009) and SB 248(2011); archival audio and video OEIB board meetings; interviews with ODE Early Learning specialists; analysis of ODE and OEIB state agency documents; and the COSA Early Learning and Full-day Kindergarten Final Report (2014). 2. Saturation – Data saturation is defined as collecting qualitative data to the point at which no new information or concepts emerge and the theories that do emerge are well supported by data (Morse, 1995, p.147). In the process of saturation, data that initially appeared diverse and disconnected form patterns and/or themes aiding the development of coherence and justification of the study’s themes. Saturation was reached through the 15 survey participants and the personal communications from Lane County superintendents of district who are currently offering full-day kindergarten. Common themes clearly emerged from both survey and personal communication responses. Additionally, Creswell (2014, p. 222) recommended that the participants in the qualitative data collection be included in the larger quantitative data sample to improve the quality of comparison between the two databases. The superintendents interviewed also participated in the survey. 3. Accounting for negative or discrepant data – As previously stated, qualitative studies construct rich understandings of the relationships between decision makers and stakeholders. As such, different perspectives from the various policy actors will not always coalesce and must be accounted for (Creswell, 2014). Patterns and themes developed through triangulation and saturation provide a framework for discussing evidence that runs counter to emerging themes. 4. Member checks or respondent validation – Member checks serve to confirm 47 the major themes and findings throughout the course of the research. Correcting errors before they are built in to the developing model and potentially subvert the analysis is key (Guba, 1981; Morse, Barrett, Mayan, Olson, & Spiers, 2008). ODE Early Learning staff, COSA staff, and superintendents served as a check on my interpretations of the interview and field observations and meaning derived from the data. 5. Clarification of researcher bias – Researcher bias occurs when the researcher has personal bias and a priori assumptions that cannot be bracketed (Onwuegbuzie & Leech, 2007). Bias may be subconsciously transferred to the participants in such a way as to influence their behaviors. Additionally, bias can influence study procedures, data collection, and analysis. During the data analysis phase, I reflected on the impact of my background as a white, educated female of privilege and as a public school district administrator on the data interpretation and identification of themes as presented in the final proposed theory. An objective analysis of data proved problematic at times, specifically regarding superintendent data. The comparison of legislative testimony, ODE interview data, and document analysis broadened my worldview regarding the complexities of education reform policy. Review of legislative testimony and ODE interview data afforded me the opportunity to consider the rationale that drives legislators as they craft policy and ODE staff as they support implementation. The passion and desire to do good works for all Oregon students was clear. Additionally, this study expanded my understanding of the challenges these bodies faced in moving districts forward. The challenges included realizing a vision of equity of outcome for all students when districts statewide hold differing views about both the definition and operationalization of equity. The districts’ responsibilities are to not simply respond in 48 frustration to policy challenges, but to honestly examine their practices and embrace a longer term view of a policy’s potential. To control for threats to external validity, descriptions of the archival audio and video recordings as well as OSBA and COSA meeting data were completed. Detailed descriptions provided a solid base for future researcher comparisons of Oregon education policymaking processes at the state and local school district levels. Onwuegbuzie and Leech (2007) recommended leaving an audit trail through maintenance of extensive documentation of records and data stemming from the study. 49 CHAPTER IV IMPLICATIONS Spillane, et al. (2002) advanced the notion that implementation failure results not because implementers reject the reform ideas but because they understand them differently than the policy makers. Scholz (1991) suggested the possibility of cooperation between policy makers and implementers where there are spaces created for both sides to exchange information in an effort to align goals and actions (p. 218). Sears and Funk (1991), on the other hand, submitted that the self-interested values and goals of a given group act as a coopting force on others who are not members. This study has the potential to inform all policy actors across a policy implementation continuum, be it federal, state or local policy. Each group of policy actors needs to consider the impact of self-oriented group interest (Sears & Funk, 1991) on their respective roles. Legislators, if they desire operationalization of their values, must consider the resources necessary for operationalization of those values when crafting legislation. Failure to do so results in the potential for failure at the implementation level. While impact studies are typically done to assess fiscal impacts within the policy implementation continuum, impact studies regarding support should be considered. Support should include the following: (a) quality and quantity of policy information required for both intermediary agents as well as implementers, (b) effective communication network to support policy sense making and input regarding needs across the continuum, and (c) necessary infrastructure and capital resources needed across the continuum. All policy has the potential for unintended outcomes. As a consequence of the electoral cycle, one policy is layered upon another as legislators are elected then replaced by the next candidate. Given this 50 environment, it is essential for legislatures to enlist state intermediaries as monitors for potential unintended consequences of policy layering. For state intermediaries, opportunities exist to expand their current role from compliance agents to true agents of change. Keedy and McDonald (2007) alluded to the potential for capacity building in districts that results in true system reform. States have the authority to perform analysis of district needs but not always the capacity, as previously discussed. If they are constrained by agency capacity, they should seek legislative support and/or partner with other advocacy groups such as COSA and OSBA to assist in that endeavor. Policy implementers, specifically districts, also have responsibilities in recognizing their own self-interested bias. The very nature of educational reform is to provide the necessary tension to be better than our current selves. If we, as a system, know we can do better, we should. Superintendents, by virtue of their position, are called to examine current practice and realize the longer term potential of reform through effective management of the short term challenges. They have a responsibility to engage and stay engaged with state intermediaries as well as legislators as part of policy work. Superintendents should strive to provide as balanced a view as possible in their advocacy work, e.g. recognizing both the policy’s potential as well as the needs to ensure effective policy implementation, articulating potential unintended consequences as clearly as possible. The study has the potential to provide information to district leaders so they can successfully respond to policy, such as full-day kindergarten, in the following ways: (a) recommend ODE and OEIB work in a collaborative manner with COSA, where ODE and OEIB communicate policy goals which COSA translates into action at the district level; 51 (b) leverage the relationships districts have with COSA and OSBA to support creation of district policies that ensure successful implementation at the school level; and (c) enable the individual school to tailor its response, while maintaining fidelity of implementation, to state policy based on the community it serves. Study participants will be offered a copy of the study upon successful completion as determined through the dissertation defense process. Reflection My interest in conducting this study was to broaden my understanding of and skill in interfacing with state and federal policymakers. My daily work as a public school superintendent requires me to be the policy sense maker and leader for the school board, community and staff. Additionally, I reflected often on the impact of my background as a white, educated female of privilege and as a public school district administrator. Objective analysis of data proved problematic at times, specifically regarding superintendent data. The comparison of legislative testimony, ODE interview data, and document analysis broadened my worldview regarding the complexities of education reform policy. Review of legislative testimony and ODE interview data afforded me the opportunity to consider the rationale that drives legislators as they craft policy and ODE staff as they support implementation. The passion and desire to do good works for all Oregon students, regardless of policy role, was clear. Additionally, this study expanded my understanding of the challenges these bodies faced in moving districts forward. The challenges included realizing a vision of equity of outcome for all students when districts statewide hold differing views about both the definition and operationalization of equity. The districts’ responsibilities are to not simply respond in frustration to policy challenges, 52 but to honestly examine their practices and embrace a longer-term view of a policy’s potential. My career has benefited from this work by reinforcing the importance of life- long learning and curiosity to effective educational leadership. I feel most fortunate to have been given the opportunity to interface with and learn from national education policy scholars. I look forward to applying my newfound skills and knowledge towards leading my organization to achieve equity of academic outcomes for all our students regardless of circumstance through effective policy implementation. 53 APPENDIX A OREGON DEPARTMENT OF EDUCATION NEWSLETTER MAY 2013 54 APPENDIX B ALIGNMENT BETWEEN RESEARCH QUESTIONS AND DATA COLLECTION INSTRUMENTS Research Question Data Collection Instruments Data Source/Participant 1. To what extent does the OEIB and ODE work collaboratively as directed in Oregon HB 3233 and Oregon SB 909 to provide information and resources that operationalize a full- day kindergarten program? A. Field Notes • OEIB Board Meetings • COSA FDK Conferences • COSA Off the Record Meetings B. Mixed forced-choice and open-ended survey • COSA FDK Workgroup members • Lane County superintendents, including Bethel, Eugene, Fern Ridge and Springfield districts C. Interviews A. Field Notes • OEIB Board Meetings (n=1) • COSA FDK Conferences (n=3) • COSA Off the Record Meetings (n=2) B. Mixed forced- choice and open ended survey • COSA FDK Workgroup Members (n=4)* • 2. Lane County Superintendents (n=11) C. Interviews • ODE Early 55 • ODE Early Learning staff • Lane County Superintendents on the COSA FDK Work Group • Non-work group Lane County Superintendents D. Document analysis of full- day kindergarten legislation; COSA Full-day Kindergarten Report; ODE Full-day Kindergarten Implementation Report, communiqués and support documents; OEIB newsletters, planning, and budget recommendation documents Learning Staff (n=2) • Lane County Superintendents on the COSA FDK Work Group (n=2) • Non-work group Lane County Superintendents (n=2) D. 1. Documents (n=30) 2. To what extent do district leaders feel prepared for full-day kindergarten implementation as a A. Mixed forced-choice and open-ended survey • COSA FDK Workgroup members A. Mixed forced-choice and open-ended survey • COSA FDK Workgroup 56 result of their relationship with ODE and OEIB? • Lane County school district superintendents B. Interviews • Lane County Superintendents on the COSA Full-day Kindergarten Workgroup • Non-work group Lane County superintendents C. Document analysis • District FDK Plans • Personal communication from other Lane County school district superintendents Members (n=4)* • Lane County Superintendents (n=11) B. Interviews • Lane County Superintendents on the COSA FDK Workgroup (n=2) • Non-workgroup Lane County superintendents (n=2) C. Documents • District FDK Plans (n=4) • Personal communication (n=3) *Two of four workgroup superintendents from Lane County 57 APPENDIX C FULL-DAY KINDERGARTEN INFORMATION SURVEY Section 1: District Information 0- 1,000 1,000- 5,000 5,000- 10,000 10,000- 20,000 Great er than 20,000 1. District Student Size 1-3 3-5 5-10 10-15 15 or more 2. Number of years experience as a public school district superintendent 3. Number of years as a superintendent of an Oregon public school district 4. Number of years as a superintendent of current district Section 2: Professional affiliation Yes No 1. Are you a member of the COSA Early Learning/Full-day Kindergarten Work Group? a. If no proceed to Section 3 b. If yes proceed to Section 4 Section 3: Full-day Kindergarten Policy Information SA A D SD NO 1. As a superintendent, I received information about the legislative intent of the full-day kindergarten legislation Oregon Senate Bill 44 (2009). Place in rank order which of the following sources you used to gather information and provide guidance for district implementation of full-day kindergarten with 0 being the least likely and 4 the most likely source. 0 1 2 3 4 2. a. Oregon Senate Bill 44 (2009) b. My State Senator c. My State Representative d. Oregon Department of Education (ODE) Deputy Superintendent Saxton e. Oregon Department of Education (ODE) Early Learning Staff Specialists f. ODE email communications and/or newsletters g. Oregon Education Investment Board (OEIB) CEO Golden 58 h. Oregon Education Investment Board (OEIB) Staff Specialists i. OEIB email communications and/or newsletters 2. j. Confederation of Oregon School Administrators (COSA) Staff k. Confederation of Oregon School Administrators Superintendents’ Early Learning Work Group l. Please list any additional sources you use to gather information and guide full- day kindergarten implementation. m. Of these additional sources, which are the most useful to you and why? 3. As a superintendent, I believe… SA A D SD NO a. ODE and OEIB provide information from both agencies that is consistent regardless of which agency provides the information. b. As a result of the information from ODE and OEIB on full-day kindergarten, I know which agency to contact for implementation questions specific to my district. c. ODE and OEIB has provided information and resources I need to lead full-day kindergarten implementation. d. ODE and OEIB adjust information and resources based on feedback from district superintendents. e. ODE and OEIB adjust information and resources based on feedback from COSA staff. f. ODE and OEIB adjust information and resources based on the COSA Early Learning Work Group Full-day Kindergarten Recommendations Report. g. ODE and OEIB adjust information and resources based on feedback from community advocacy and/or association groups, e.g. Chalk Board Project, OEA, Stand For Children, Early Learning Hubs. h. My district has the highly qualified 59 kindergarten teachers and facilities to implement full-day kindergarten. Section 4: Policy Information COSA Early Learning Work Group Superintendents SA A D SD NO 1. As a superintendent, I received information about the legislative intent of the full-day kindergarten legislation Oregon Senate Bill 44 (2009). Place in rank order which of the following sources you used to gather information and provide guidance for the COSA Work Group on Full-day Kindergarten with 0 being the least likely and 4 the most likely source. 0 1 2 3 4 2. a. Oregon Senate Bill 44 (2009) b. My State Senator c. My State Representative d. Oregon Department of Education (ODE) Deputy Superintendent Saxton e. Oregon Department of Education (ODE) Early Learning Staff Specialists f. ODE email communications and/or newsletters g. Oregon Education Investment Board (OEIB) CEO Golden h. Oregon Education Investment Board (OEIB) Staff Specialists i. OEIB email communications and/or newsletters j. Confederation of Oregon School Administrators (COSA) Staff k. Please list any additional sources you use to gather information and guide the work of the COSA Full-day Kindergarten Work Group. l. Of these additional sources, which are the most useful to you and why? 3. As a superintendent, I believe… SA A D SD NO a. ODE and OEIB provide information from both agencies that is consistent regardless of which agency provides the information. b. As a result of the information from ODE and OEIB on full-day kindergarten, I know which agency to contact for 60 implementation questions specific to my district. c. ODE and OEIB has provided information and resources I need to lead full-day kindergarten implementation. d. ODE and OEIB adjust information and resources based on feedback from district superintendents. e. ODE and OEIB adjust information and resources based on feedback from COSA staff. f. ODE and OEIB adjust information and resources based on the COSA Early Learning Work Group Full-day Kindergarten Recommendations Report. g. ODE and OEIB adjust information and resources based on feedback from community advocacy and/or association groups, e.g. Chalk Board Project, OEA, Stand For Children, Early Learning Hubs. SA A D SD NO h. My district has the highly qualified kindergarten teachers and facilities to implement full-day kindergarten. APPENDIX D INTERVIEW QUESTIONS: SUPERINTENDENTS 1. One of the roles of a state department of education is to provide guidance and oversight to districts regarding implementation of state education legislation and policy. a. Is the information ODE and OEIB communicates to districts useful to you as you plan for implementation of full-day kindergarten? Why or why not? b. In what ways does ODE and OEIB solicit feedback from districts regarding the efficacy of the information they provide to districts regarding full-day kindergarten? c. In what ways does ODE and OEIB adjust the information and support to districts based on feedback from districts? Please provide specific examples from your district’s perspective. 2. My survey of superintendents rated COSA as the first source for information concerning information for planning for full-day kindergarten, ahead of information and communications from ODE and OEIB staff or their chief executive officers. a. What actions on COSA’s part accounts for this data? b. What role does COSA play in providing information to your district regarding the intent of full-day kindergarten legislation? 61 Question for Superintendents Who Were Not Members of the COSA Kindergarten Work Group 3. My survey of superintendents who were not involved in the COSA Full-day Kindergarten Work Group rated the information from the work group as a significant source of information for planning full-day kindergarten, ahead of information and communications from ODE and OEIB staff or their chief executive officers. What information in the work group report do you believe accounts for this data? Questions for Superintendents Who Were Members of the COSA Kindergarten Work Group 4. My survey of superintendents who were members of the work group rated information from the work group report as a more significant source of information for planning full-day kindergarten, ahead of information and communications from ODE and OEIB staff or their respective chief executive officers. What experiences from the work group meetings and/or report do your believe accounts for this data? 5. What role did the COSA Work Group play in providing information to ODE and/or OEIB regarding district capacity for full-day kindergarten? 6. What role did the COSA Work Group play in providing information to legislators regarding district capacity for full-day kindergarten? 7. Is there anything I haven’t asked that you feel would help me understand information flow around preparing to implement full-day kindergarten? Is there anyone else you believe I should interview regarding this topic? APPENDIX E INTERVIEW QUESTIONS: ODE STAFF 1. One of the roles of a state education department is to provide guidance and oversight to districts regarding implementation of state education legislation and policies. For the following questions, please provide examples of effective methods as well as examples where there were challenges with the methods. a. How does ODE determine what information and support are needed by districts to ensure a district may successfully plan for implementation of any new legislation and/or policy? b. What methods are used by ODE to deliver information and/or support to districts? c. How does ODE monitor the effectiveness of the information and/or support provided to districts? d. How does ODE solicit feedback from districts regarding the efficacy of the information and/or support in supporting district planning for implementation of a new policy? e. What adjustments in messaging to districts, if any, are made based on district feedback? 62 2. With the passage of 40-40-20 legislation, ODE and OEIB were directed to work collaboratively in support of the state education strategic initiatives. a. What have been the successes of the relationship between ODE and OEIB and what have been the challenges when providing information to districts about strategic initiatives? b. What information does ODE coordinate with OEIB in communications to districts as they plan for implementation of strategic investments? Please provide specific examples. c. What information does ODE coordinate with OEIB in communications to the legislature regarding district capacity for planning and implementation of strategic investments? Please provide specific examples. 3. How has the information and support to districts from ODE regarding full-day kindergarten planning changed as a result of OEIB’s “Pathway to Kindergarten Readiness and 3rd Grade Reading” initiative? 4. Anything I haven’t asked that you feel would help me understand information flow around preparing to implement full-day kindergarten? Is there anyone else you believe I should interview concerning this topic? 63 APPENDIX F SURVEY DATA 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 APPENDIX G ATLAS.TI QUOTATION OUTPUT 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 APPENDIX H RESEARCH SURVEY QUESTION MAP Research Question Survey Items Addressing Question RQ 1: To what extent does the OEIB and ODE work collaboratively as directed in Oregon HB 3233 to provide information and resources that operationalize a full-day kindergarten program? 3.2.d.,3.2.e.,3.2.f.,3.2.g.,3.2.h.,3.2.i.,3.2.l.,3.2.m. 3.3.a, 3.3.b., 3.3.c., 3.3.d., 3.3.e., 3.3.f, 3.3.g. 4.2.d., 4.2.e., 4.2.f., 4.2.g., 4.2.h., 4.2.i. 4.3.a., 4.3.b., 4.3.c., 4.3.d., 4.3.e, 4.3.f., 4.3.g. RQ 2: To what extent do district leaders feel prepared for full-day kindergarten implementation as a result of their relationship with ODE and OEIB? 1.1., 1.2.,1.3.,1.4. 2.1. 3.1. 3.2.a, 3.2.b.,3.2.c.,3.2.j,3.2.k.,3.2.l.,3.2.m. 3.3.h. 4.1. 4.2.a.,4.2.b.,4.2.c.,4.2.j.,4.2.k,4.2.l. 4.3.h. 96 APPENDIX I OEIB PROPOSED SCHOOL FUNDING PRIORITIES 2015-17 Pathway to Kindergarten Readiness and 3rd Grade Reading As#we#build#an#outcome1based#budget#in#a#P120#system,#we#have#the#opportunity#to#invest#where#we#will# get#the#greatest#return.#Healthy#babies,#stable#and#attached#families,#and#quality#childcare#and#early# learning#experiences#are#what#is#best#for#Oregon.#More#students,#especially#students#of#color#and# students#from#poverty,#must#begin#Kindergarten#ready#to#learn#both#academically#and#socially.#Closing#this# opportunity#gap#will#help#ensure#higher#achievement#for#every#student#in#grades#K13.#Students#who#are# reading#at#grade#level#at#3rd#grade#are#four#times#less#likely#to#drop#out;#reading#is#the#key#skill#students# need#to#access#advanced#content,#continue#to#post1secondary#education,#and#connect#their#education#to#a# career.#In#order#to#reach#the#key#outcome#of#proficient#3rd#grade#reading#for#95%#of#Oregon#students#by# third#grade,#a#multi1faceted#approach#is#necessary.#For#the#near#term,#immediate#improvements#in#district# elementary#schools#and#systems#will#support#the#students#already#in#school.#These#improvements#include# a#full#school#day#for#Kindergarten,#a#research1based#reading#program#for#all#elementary#schools,# partnerships#with#community#groups#better#equipped#to#provide#wrap1around#services#and#additional# instructional#time,#and#improved#educator#performance.#For#the#long#term,#a#targeted#intervention#in# early#learning#will#increase#the#percentage#of#students#entering#Kindergarten#ready#to#learn.#The#creation# of#early#learning#systems,#grounded#in#the#community#and#involving#all#partners,#leverages#collective# impact#to#produce#family#stability,#health,#and#school1readiness.# ! Strategy! Investment! Outcome! 1.!PreK63!Reading! • All#Day#Kindergarten#(220#M)# • PreK#to#Grade#3#Literacy#(180#M)# • Dual#Language#Progress#Monitoring/Spanish#Language# Assessments#(0.1#M)# # $400.1#M# 95%#of#students#proficient#at# 3rd#grade#reading# 2.!Early!Learning! • Aligned#Home#Visiting#(10#M)# • Early#Intervention/Early#Childhood#Special#Education#(15# M)# • Employment#Related#Daycare#(55#M)# • Targeted#Pre1school#strategy#(30#M)# • Early#Learning#Hubs#(20#M)# • K1Readiness#Partnerships#and#Innovation#(5#M)# # $135#M# 8%#increase#in#K#readiness# 3.!Network!for!Quality!Teaching! • Culturally#Responsive#Teaching#Practices#(4#M)# • K112#Mentoring#(11#M)# • Expansion#of#School#District#Collaboration#(17#M)# • Support#for#Low#Performing#Districts#(5.4#M)# # $37.4#M# 95%#of#students#proficient#at# 3rd#grade#reading# 97 REFERENCES CITED Babbie, E. R. (2010). The practice of social research. Belmont, CA: Cengage Learning. Bowen, G. A. (2009). Document analysis as a qualitative research method. Qualitative Research Journal, 9, 27-40. Braun, D. (1993). Who governs intermediary agencies? Principal-agent relations in research policy-making. Journal of Public Policy, 13, 135-162. Retrieved from. http://www.jstor.org/stable/4007501 Coleman, J. S. (1994). Foundations of social theory. Boston, MA: Harvard University Press. Conley, D. (2003). Who governs our schools? New York, NY: Teachers College Press. Conley, D. & Goldman, P. (1995). Reactions from the field to state restructuring legislation. Educational Administration Quarterly, 31, 512-538. Conley, D., & Goldman, P. (2000). Half full, half empty? Educator responses over time to state-level systemic reforms. International Journal of Educational Reform, 9, 249-260. Creswell, J. W. (2014). Research design: Qualitative, quantitative, and mixed methods approaches. Sage Publications, Incorporated. Creswell, J. W. & Plano Clark, V. L. (2011). Designing and conducting mixed methods research. Thousand Oaks, CA: SAGE Publications, Inc. Dahill-Brown, S. E., & Lavery, L. (2012). Implementing federal policy: Confronting state capacity and will. Politics & Policy, 40, 557-592. Eisenhardt, K. M. (1989). Agency theory: An assessment and review. Academy of Management Review, 14, 57-74. Retrieved from http://www.jstor.org/stable/258191 Elmore, R. F. (1980). Backward mapping: Implementation research and policy decisions. Political Science Quarterly, 94, 601–616. Elmore, R. F. (1996). Getting to scale with good educational practice. Harvard Educational Review, 66, 1–26. Fink, A. (2013). How to conduct surveys: A step-by-step guide. (5th ed.). Los Angeles, CA: Sage. 98 Fowler, F. (2013). Policy studies for educational leaders: An introduction. Upper Saddle River, NJ: Pearson Education. Fuchs, S. (2001). Beyond agency. Sociological Theory, 19, 24-40. Fullan, M. (1996). Turning systemic thinking on its head. The Phi Delta Kappan, 74, 420–423. Fullan, M., & Miles, M. (1992). Getting reform right: What works and what doesn't. Phi Delta Kappan, 73, 744-752. Georgia Department of Education, (2010). Investing in educational excellence IE2 partnership contracts. Retrieved from http://www. gadoe.org/External-Affairs- and-Policy/Policy/Pages/IE2.aspx Golafshani, N. (2003). Understanding reliability and validity in qualitative research. The Qualitative Report, 8, 597-607. Goldman, P., & Conley, D. T. (1995, April). Systemic School Reform in Oregon: Can it be legislated? Paper presented at the annual meeting of the American Educational Research Association, San Francisco, CA. Retrieved from http://eric.ed.gov/?id=ED383103 Guba, E. G. (1981). Criteria for assessing the trustworthiness of naturalistic inquiries. ECTJ, 29, 75-91. Hirsch, P., Michaels, S., & Friedman, R. (1987). “Dirty hands” versus “clean models.” Theory and Society, 16, 317-336. Honig, M. I. (2004). Where’s the “up” in bottom-up reform? Educational Policy, 18, 527-561. doi: 10.1177/0895904804266640 Ingram, H., & Schneider, A. (1990). Improving implementation through framing smarter statutes. Journal of Public Policy, 10(1), 67-88. Retrieved from http://www.jstor.org/stable/4007346 Keedy, J. L., & McDonald, D. H. (2007). The instructional capacity building role of the state education agency: Lessons learned in Kentucky with implications for No Child Left Behind. Planning and Changing, 38, 131–147. Kipo, D. D. (2013). Agency-structure relation in social sciences: Reflections on policy implementation. Asian Social Science, 10, 18-26. doi: 10.5539/ass.v10n2p18 Louis, K., Febey, K., Gordon, M., Meath, J., & Thomas, E. (2006). Educational leadership in the states: A cultural analysis. St. Paul, MN: University of Minnesota. 99 Louis, K. S., Thomas, E., Gordon, M. F., & Febey, K. S. (2008). State leadership for school improvement: An analysis of three states. Educational Administration Quarterly, 44, 562-592. doi: 10.1177/0013161X08323858 McDonnell, L. M. & Elmore, R. F. (1987). Getting the job done  : Alternative policy instruments. Educational Evaluation and Policy Analysis, 9, 133–152. Retrieved from http://www.jstor.org/stable/1163726 McPhee, D. M., & Bronstein, L. (2002). Constructing meaning: Strengthening the policy- practice link. Social Work Education, 21, 651-662. Melton, K. (2011, December 02). Gov. John Kitzhaber plans a powerful Oregon education board, connecting school funding to performance. OREGONLIVE. Retrieved from: http://blog.oregonlive.com/politics_impact/print.html?entry=/2011/02/gov_jo hn_kitzhaber_plans_to_cr.html Miller, G. J. (2005). The political evolution of principal-agent models. Annual Review of Political Science, 8, 203-225. doi: 10.1146/annurev.polisci.8.08213.104840 Mitnick, B. M. (1992). The theory of agency and organizational analysis. Ethics and agency theory: An introduction. New York, NY: Oxford University Press. Morse, J. M. (1995). The significance of saturation. Qualitative Health Research, 5, 147- 149. Morse, J. M., Barrett, M., Mayan, M., Olson, K., & Spiers, J. (2008). Verification strategies for establishing reliability and validity in qualitative research. International Journal of Qualitative Methods, 1(2), 13-22. Mulhall, A. (2002). In the field: Notes on observation in qualitative research. Journal of Advanced Nursing, 41, 306-313. Muth, R. (1984). Toward an integrative theory of power and educational organizations. Educational Administration Quarterly, 20, 25-42. doi: 10.1177/0013161X84020002003 North Carolina State Board of Education, (2012). Education in North Carolina: A great public education system for a great state. Retrieved from http://stateboard.ncpublicschools.gov/resources/north-carolina-ambassador- resources/vision-report.pdf/view Oregon Department of Education. Rob’s Blog. ODE Education Update, May 2013. http://www.ode.state.or.us/pipeline/education-update-may-2013.pdf 100 Oregonian Editorial Board (2015, April 12). What partisan squabble over K-12 funding is really all about. The Oregonian. Retrieved from: http://www.oregonlive.com/opinion/index.ssf/2015/04/the_squabble_over_the _k-12_bud.html OEIB, (2013). Prepared students = prosperous Oregon. Retrieved from http://education.oregon.gov/Document/OEIB_brochure_NSIF_lowres.pdf Oregon ESEA Flexibility Request (2012). Retrieved from the Oregon Department of Education website http://www.ode.state.or.us/initiatives/nclb/docs/oregons_completed_esea_flex ibility_request_abridged.pdf Oregon House Bill 3233, HB 3233, 77th Oregon Legislative Assembly (2013). Retrieved from https://olis.leg.state.or.us/liz/2013R1/Measures/Overview/HB3233 Oregon Senate Bill 44, SB 44, 75th Oregon Legislative Assembly (2009). Retrieved from https://olis.leg.state.or.us/liz/2009R1/Downloads/MeasureDocument/SB44/En rolled Oregon Senate Bill 248, SB 248, 76th Oregon Legislative Assembly (2011). Retrieved from http://olis.leg.state.or.us/liz/2011R1/Measures/Overview/SB248 Oregon Senate Bill 253, SB 253, 76th Oregon Legislative Assembly (2011). Retrieved from http://olis.leg.state.or.us/liz/2011R1/Measure/Overview/SB253 Oregon Senate Bill 909, SB 909, 76th Oregon Legislative Assembly (2011). Retrieved from http://www.ode.state.or.us/superintendent/priorities/2011-august-16-sb- 909-b.pdf Oregon University System (2011). From goal to reality 40-40-20: A report on strategies to meet Oregon’s 40-40-20 education goals. Retrieved from http://www.oregon.gov/gov/oeib/docs/nnousreport.pdf Onwuegbuzie, A. J., & Leech, N. L. (2007). Validity and qualitative research: An oxymoron? Quality & Quantity, 41, 233-249. doi: 10.1007/s11135-006-9000- 3. Perrow, C. (1986). Economic theories of organization. Theory and Society, 15, 11-45. Retrieved from http://www.jstor.org/stable/657174 Pogrow, S. (1996). Reforming the wannabe reformers: Why education reforms almost always end up making things worse. Phi Delta Kappan, 77, 656-663. Retrieved from http://www.jstor.org/stable/20405675 101 Sabatier, P. A. (1986). Top-down and bottom-up approaches to implementation research: a critical analysis and suggested synthesis. Journal of Public Policy, 6, 21-48. Retrieved from http://www.jstor.org/stable/3998354 Scholz, J. T. (1991). Cooperative regulatory enforcement and the politics of administrative effectiveness. American Political Science Review, 85, 115-136. Retrieved from http://www.jstor.org/stable/1962881 Scribner, J. D., Aleman, E., & Maxcy, B. (2003). Emergence of the politics of education field: Making sense of the messy center. Educational Administration Quarterly, 39, 10-40. doi: 10.1177/0013161X02239759 Sears, D. O., & Funk, C. L. (1991). The role of self-interest in social and political attitudes. Advances in Experimental Social Psychology, 24, 1-91. Shapiro, S. P. (2005). Agency theory. Annual Review of Sociology, 31, 263-284. Retrieved from http://www.jstor.org/stable/29737720 Spillane, J. P. (2004). Standards deviation: How schools misunderstand education policy. Cambridge, MA: Harvard University Press. Spillane, J. P., Reiser, B. J., & Reimer, T. (2002). Policy implementation and cognition: Reframing and refocusing implementation research. Review of Educational Research, 72, 387-431. Retrieved from http://www.jstor.org/stable/3515992 Stake, R. E. (1978). The case study method in social inquiry. Educational Researcher, 7(2), 5-8. Retrieved from http://www.jstor.org/stable/1174340 Waterman, R. W., & Meier, K. J. (1998). Principal-agent models: an expansion? Journal of Public Administration Research and Theory, 8, 173-202. Retrieved from http://jpart.oxfordjournals.org Wood, B. D., & Waterman, R. W. (1993). The dynamics of political-bureaucratic adaptation. American Journal of Political Science, 37, 497-528. doi: 10.2307/2111