IJEPL Volume 11(9) 2016 Influential Spheres: Examining Actors’ Perceptions of Education Governance Michael Thier, MAT, University of Oregon Joanna Smith, PhD, University of Oregon Christine Pitts, MAEd, University of Oregon Ross Anderson, University of Oregon Abstract Many layers of education governance press upon U.S. schools, so we sep- arated state actors into those internal to and those external to the system. In the process, we unpacked the traditional state–local dichotomy. Using interview data (n=45) from six case-study states, we analyzed local leaders’, state-internal actors’, and state-external players’ perceptions of implementation flexibility and hindrances across several policy areas. We observed how interviewees’ spheres of influence linked to which policy areas they viewed as salient or not, and their relative emphases on whoand whatwithin state education systems contributed to implementation flex- ibility and/or hindrances, and how these factors played out. We found important dif- ferences by sphere: the local sphere produced the most coherent findings, and state-internal was least coherent. We discuss implications for education governance research, applications for practitioners and policymakers, and a methodological con- tribution. Keywords Education governance, Education policy, Research methods, State and local perspectives Michael Thier, Joanna Smith, Christine Pitts, & Ross Anderson (2016). Influential Spheres: Examining Actors’ Perceptions of Education Governance. International Journal of Education Policy & Leadership 11(9). URL: http://journals.sfu.ca/ijepl/index.php/ijepl/article/view/682 IJEPL is a joint publication of PDK International, the Faculty of Education at Simon Fraser University and the College of Education and Human Development at George Mason University. By virtue of their appearance in this open access journal, articles are free to use, with proper attribution, in educational and other non-commercial settings 90 days after initial publication. Copyright for articles published in IJEPL is retained by the authors. More information is available on the IJEPL website: http://www.ijepl.org IJEPL11(9) 2016 Introduction Thier, Smith, Pitts, A multi-layered, complex system governs public education in the United States, in- & Anderson cluding (a) an increasingly prominent federal governance role, (b) numerous state- Perceptions level organizations, and (c) regional and local school districts, all creating policies of Education Governance for school personnel to implement (McGuinn & Manna, 2013; Torres, Zellner, & Erlandson, 2008). Different levels create overlapping spheres of influence with “ill- defined responsibilities and often conflicting interests” (Finn & Petrilli, 2013, p. 21). Furthermore, U.S. public schools answer to “multiple sources of funding and nu- merous masters who sometimes possess conflicting priorities and demand incongru- ous results” (McGuinn & Manna, 2013, p. 7). Researchers typically conceptualize these overlapping spheres of influence, which Finn and Petrilli (2013) characterize as “too many cooks in the education kitchen” (p. 32), within a state–local dichotomy. But state institutions make up a “complex web” of education governance (McGuinn & Manna, 2013, p. 5), including groups inside and outside of formal governance structures. Internal organizations include governors, state legislatures, state courts, state boards of education (in 47 of 50 states), and state education agencies, which are led by Chief State School Officers and which also contain numerous divisions that oversee schools’ adherence to various state and federal policies. External organizations include curriculum and test developers, unions, business groups, and advocacy groups influencing policy and governance by lobbying legislatures, rallying parents and communities, and forming coalitions to push for reforms. In the United Kingdom, Ball and Exley (2010) theorized a rise in the status of external players, leading to “polycentric” education governance, whereby multiple agents—governmental and not—contribute to policy production. Separating internal and external institutions as distinct spheres within state-level governance can aid in the examination of both structural roadblocks to and enablers of educational reform in the local sphere. At the same time, the U.S. Constitution’s reservation of authority for the states has created wide variations both across (Torres et al., 2008) and within states (McGuinn & Manna, 2013), resulting in differing roles across jurisdictions. These spheres of influence lead to differing actions and re- actions by local school leaders, due to a hierarchy in American public schools that has “produced a compliance culture that stifles the ability and willingness of school teachers and leaders to improve school practice organically or to faithfully or effec- tively implement external reforms” (McGuinn & Manna, 2013, p. 7). Local leaders’ capacity or willingness to implement policies harkens back to Lipsky’s (1980) seminal work, Street level bureaucrats, in which he argued that front- line workers (e.g., teachers) wield considerable discretion in the day-to-day imple- mentation of policies, constrained by limited resources. In this vein, Fowler (2013) argues that “many official policies are never implemented at all, and many others are implemented only partially or incorrectly” (p. 241) due to lack of will and/or ca- pacity of the individuals tasked with implementation. Fowler notes that “imple- menters perform best when they are receiving messages about new policies from multiple sources in the environment, and not just from their school district or the state department of education” (p. 251). However, given the multiple cooks in the education governance kitchen, policy implementers also receive conflicting messages 2 IJEPL11(9) 2016 that might obstruct implementation. As a result, policymakers across spheres share Thier, Smith, Pitts, the “dilemma of improving schools while maintaining morale” (Torres et al., 2008, & Anderson p. 7), especially in an era when policy development is “mostly in the hands of policy Perceptions elites,” and practitioners’ perceptions receive less attention (p. 2). of Education Governance The current study stems from a program of inquiry that examined how states with different governance arrangements approach a range of policy areas: (a) curricu- lum adoption; (b) teacher evaluations; (c) teacher licensure/certification; (d) instruc- tional materials adoption; (e) interventions for chronically underperforming schools (i.e., takeover of schools/districts); (f) taxation/budgeting authority; and (g) overall education governance. The current study examines how actors from different spheres of influence vary in their interpretations of how their state systems contribute to im- plementation flexibility and/or hindrances. Our program of inquiry follows theory from Brewer and Smith (2008), who posited that understanding state education gov- ernance involves examining a state system’s what,who,and how(see Figure 1). What includes the necessary set of functions that require organization in the context of sys- tem goals (e.g., What are the necessary functions to be accomplished? What programs or policies will schools and districts implement and/or emphasize?). Whorecognizes the institutions and individuals responsible for fulfilling each of the what functions, including various organizations and stakeholders at state (internal and external) and local spheres. Howasks whether the whatfunctions operate from mandates, induce- ments, capacity-building, or system-changing (see McDonnell & Elmore, 1987). Figure 1: The what, who, and howof state education governance WHATare the goals of the system in terms of: • Structure and Organization • Fianance and Business Services • Human Resources/Personnel • Education Programs? WHOis best situated to carry out the tasks necessary to meet these goals? Think about institutions and individuals at the various levels of the system (e.g., Governor, Legislature, State Board, State Superintendent, State Department, District Boards, County Offices of Education, Principals and Teachers) HOWshould these institutions or individuals best induce others to implement policy? What mix of the following is best suited to meet the goals: • Mandates • Inducements • Capacity-Building • System-Changing Source: Adapted from Brewer & Smith (2008) Brewer and Smith’s what-who-how framework informed the coding schema that we employed to capture the following common themes across policy areas: excep- 3 IJEPL11(9) 2016 tions, innovations, frequency, flexibility, mixed opinion, positive enablers, and hin- Thier, Smith, Pitts, drances. Inductively, we focused the current study on flexibility and hindrances after & Anderson observing coding patterns that highlighted differences among interviewees from the Perceptions three spheres: local leaders, state-internal actors, and state-external players. In line of Education Governance with the notion of working hypotheses in qualitative research—“hypotheses that re- flect situation-specific conditions in a particular context” (Merriam, 2009, p. 225)— we expected spheres to yield different perspectives about what does and does not work within states. Therefore, we asked: 1. Do local leaders, state-internal actors, and state-external players find similar policy areas salient? 2. Do local leaders, state-internal actors, and state-external players cohere in the way they emphasize the what, who, or how of pol- icy areas? We define saliencefor the current study as the proportion of interviewees whose comments received a given code. We define emphasis as which of the three dimen- sions in Brewer and Smith’s 2008 framework (i.e., what, who, or how) resonated most within each sphere. After re-examining our data for salience and emphasis, we characterized spheres for coherence(i.e., similarities or agreement within the sphere). Method First, we describe the sources and procedures we used to collect interview data from policy actors across six case-study states—California, Indiana, Kentucky, Ohio, Oregon, and Tennessee—for the overall program of inquiry. Second, we explain the analytical processes used in the current study. Data collection and sources In the overall program of inquiry, Smith, Thier, Gasparian, Anderson, Shen, and Pitts (2015) analyzed education policies across the 50 states and the District of Columbia along three facets of education governance: (a) level of control, (b) distribution of authority, and (c) degree of participation.1To sample six case-study states based on maximum variation (Patton, 2002), we used our 2015 findings, employing indicators of states’ roles in education governance. For example, Tennessee scored high on level of control; Oregon scored low. Our other four states scored in the middle of the na- tional distribution. Also, we sought distributional variety among states’ scores on de- gree of participation, with indicators of whether the state board was required to include an array of stakeholders.2California and Indiana scored low on participation, Kentucky and Oregon scored in the middle, and Tennessee and Ohio scored high. In Appendix A, we provide the semi-structured interview protocol we used to elicit responses on the range of policy areas. We examined three policy areas per state; each state protocol concluded with overall education governance questions, which were common across states. The protocol featured scenarios that pertained to policies and implementation processes, not specific policies themselves. Within each state, we used stratified snowball sampling (Patton, 2002) to select interview participants (n = 45). First, we interviewed State Board of Education mem- bers, closing interviews by requesting nominations of organizations, districts, or pol- 4 IJEPL11(9) 2016 icy actors that particularly opposed and/or supported the policy areas covered in the Thier, Smith, Pitts, interviews. Nominated participants included individuals with state-internal roles & Anderson (e.g., Chief State School Officers, Department of Education administrators, or State Perceptions Board members) and state-external roles (e.g. union leaders, members of education- of Education Governance related councils, advocacy groups, and business association leaders). We concluded state-internal and state-external interviews by seeking nominations for additional interviewees at state and local spheres. We sought local participants from a range of districts (e.g., low-performing/high-performing, urban/rural, high poverty/affluent, high and low proportions of English learners) and included local school board mem- bers and district superintendents or other administrators. During academic year 2013–2014, we interviewed 6–9 local leaders, state- internal actors, and state-external players per state (see Table 1). As Merriam (2009) suggests, a “small, nonrandom, purposeful sample is selected precisely because the researcher wishes to understand the particular in depth, not to find out what is gen- erally true of the many” (p. 224). We opted for depth over breadth of interviews, spending around an hour with each interviewee to gain nuanced perspectives from stakeholders with various roles within each state rather than seeking a generalizable sample that would be representative of the states selected. Our sample enables us to offer extrapolations—“modest speculations on the likely applicability of findings” (Patton, 2002, p. 584)—from the data, which might be applicable to other situations, namely other states experiencing similar spheres of influence in which local leaders, state-internal actors, and state-external players all contribute to education governance. Table 1: Interviewees by state and sphere of influence State Local State-internal State-external California 2 2 2 Indiana 4 3 2 Kentucky 3 1 2 Ohio 4 4 1 Oregon 4 2 1 Tennessee 4 3 1 Total 21 15 9 Thematic coding and analysis To achieve consistency for the program of inquiry, the researcher with the most cod- ing experience completed all initial data coding after the entire team piloted and re- fined the code list. When reviewing the primary analysis, the lead author of the current study detected several examples in which clusters of codes revealed two high-salience spheres and one low-salience sphere, or one high-salience sphere and two low-salience spheres. For example, 12 of 22 local leaders (55%) had at least one quotation coded for curriculum hindrances. In contrast, 4 of 15 state-internal actors (27%) and 1 of 9 state-external players (11%) had at least one quotation coded for curriculum hindrances. By this calculus, we determined curriculum hindrances to present greater salience among local leaders than among either their state-internal 5 IJEPL11(9) 2016 or state-external counterparts. Therefore, the secondary coder re-analyzed the data Thier, Smith, Pitts, for both salience and emphasis (i.e., what, who, or how). Topics that emerged from & Anderson secondary coding included: (a) curriculum hindrances, (b) intervention hindrances, Perceptions (c) budget hindrances, (d) teacher evaluation flexibility, (e) curriculum flexibility, (f) of Education Governance teacher certification and hiring hindrances, and (g) overall education governance hindrances. In Table 2, we report the salience and dimension of greatest emphasis for each code by sphere of influence. Table 2: Salience and emphasis for policy actors by sphere of influence Local State-internal State-external (n= 21) (n= 15) (n= 9) Topic Sal. Emph. Sal. Emph. Sal. Emph. Curriculum hindrances 55% how 27% none 11% none Intervention hindrances 55% what, how 33% none 11% none Budget hindrances 68% how 33% what, how 44% how Teacher evaluation 14% how 7% what, how 33% who, how flexibility Curriculum flexibility 59% how 50% who, how 33% who Teacher certification 18% none 13% none 44% what and hiring hindrances Overall education 64% who 27% none 67% who governance hindrances Note: Sal. = Salience; Emph. = Emphasis. We excluded codes if salience did not vary across spheres by more than 11%, establishing that threshold because our least-populated sphere (state-external) contained nine actors. We could not justify inclusion or exclusion due to presence or absence of a single member in that sphere, a swing of +/- 11%. Findings Analyzing for salience by sphere showed differences in how local leaders, state-in- ternal actors, and state-external players saw policy areas as hindrances or flexible opportunities with respect to their states’ approaches to education governance. Local leaders emphasized processes (how) in 5 of 7 topics, despite coming from disparate states. Overall, though state-internal and state-external interviewees revealed differ- ences in salience, neither group cohered in their dimension of emphasis to the extent that local leaders’ interview responses demonstrated. In this section, we organized findings to show how our salience indicator revealed (a) local leaders to contrast with state-internal actors and state-external players in three instances, (b) state-ex- ternal players to contrast with local leaders and state-internal actors in three in- stances, and (c) state-internal actors to contrast with local leaders and state-external players in one instance. Local focus Local leaders were the most coherent sphere, especially when discussing hindrances around curriculum, budgets, and interventions for chronically underperforming schools. 6 IJEPL11(9) 2016 Curriculum hindrances Thier, Smith, Pitts, Local leaders across 5 of 6 case-study states described practical challenges around & Anderson curriculum (how), a topic that 16 of 22 local leaders referenced (73%). Nine of the Perceptions 16 who addressed howspecified state processes as the primary culprit for curriculum of Education Governance challenges: several cited needs to increase districts’ capacities to vet and/or upgrade curricula. For example, in both Kentucky and Indiana, local leaders characterized state-led implementation of the Common Core State Standards (CCSS) as unfunded legislative mandates. A local leader in Indiana questioned the state’s inducements for professional development, saying, “That’s ridiculous to think that we’re going to send 1,700 of our teachers throughout a year to professional development that is not offered here.” In another case, a superintendent of a small district in Oregon wished his staff could capitalize on state laws that permit local choice in curriculum adoption. Instead, his district conforms to state expectations because he lacks the resources to inform choices locally. Another Oregon superintendent identified a flaw in the process for adding new curriculum to the state’s adoption list: publishers must fund reviews that the state department of education conducts. Otherwise, their cur- ricula cannot be vetted, a process that empowers large publishers to crowd out smaller competitors who might better meet certain schools’ needs. Two local leaders diverged from their sphere’s trend toward favoring local over state control of curriculum. Both actors identified processes in which local control created or perpetuated capacity gaps. An actor in Tennessee praised the statewide capacity-building process of employing curriculum specialists in each district “to very deeply integrate our culture into the curriculum transition process” toward CCSS. Instead of noting how-type challenges, as did most other local leaders, he identified “political roadblocks” around CCSS. In Kentucky, where curricular deci- sions belong exclusively to school councils—typically, 2-3 practitioners (i.e., teachers and/or administrators) and 2-3 parents from the school community—a local leader in a high-mobility district bemoaned one outgrowth of the councils’ highly localized authority: If a student is at one of my elementary schools under one reading curriculum [and] moves to another elementary school … just across the street, under school governance, [that student] could change curricula almost entirely from one school to the next within the same district. I think that brings its own level of complexity that sometimes is lamentable. In contrast to the high salience of curriculum hindrances among local leaders, this topic appeared much less salient for state-internal actors and state-external play- ers; only 5 of 24 non-local interviewees referenced curriculum hindrances (21%). For those state-internal actors that did identify curriculum hindrances, their percep- tions revealed less cohesion than we found among local leaders. For example, a state- internal actor in Kentucky detailed political conflict over potential adoption of the Next Generation Science Standards, specifically regional controversies around teach- ing evolution and climate change. That conflict has been conflated with CCSS adop- tion due to popular claims about CCSS as a federal mandate that dangled Race to 7 the Top funds as its carrot. Focusing on who had a crucial role in curriculum hin- IJEPL11(9) 2016 drances, one state-internal actor in Ohio called for school board members to “do Thier, Smith, Pitts, & Anderson their homework” upon receiving superintendents’ curriculum recommendations. Otherwise, board members should expect “mud on their faces.” Perceptions of Education Governance Intervention hindrances Though local leaders seemed equally emphatic about whatand how when discussing intervention hindrances, the topic overall held a high degree of salience (55%). Meanwhile, intervention hindrances were less salient among state-internal actors (33%) and state-external players (11%), who demonstrated no coherent emphases. Contrasts observed in this topic showed the state-internal sphere, in particular, to be less unified. Local leaders consistently depicted state interventions as riddled with problems or as having little effect. One local Indiana leader doubted whether a state-hired ex- ternal consultant could be of any permanent benefit to a weak school leader, because weak leaders struggle to attract and retain good teachers, resulting in “a mix that doesn’t work” even with expenditures in the “millions of dollars.” In Ohio, a local leader presented a more neutral view of state interventions. He spoke broadly of suc- cesses and failures when the state redesigned six low-performing elementary schools, revising curriculum and replacing at least half the faculty and staff: “We even take out the clerical and the lunchroom lady and everyone and start over.” Yet, he ex- pounded on the difficulty of finding quality institutions to sponsor charter operators as takeover options for chronically under-performing schools, often due to some charter operators’ financial mismanagement. He described the annual problem of charter schools closing midyear: “Children arrive to find the doors padlocked,” which compels those students to return to their zoned schools after low-performing operators are allowed to “hire your husband’s cleaning company, your daughter as the secretary, and you pay them six-figure salaries.” An Indiana local leader reported that state law restricts districts’ abilities to use professional development as an inter- vention. The state requirement of 180 days of instruction in the year, combined with the inability to count half days in that total as other states allow, makes it impossible to offer professional development during a school day. The interviewee cited research suggesting that professional development “should be job-embedded, connected to their work time. We can’t do that without adding days under the calendar, which then adds more money.” State-internal actors were less consistent in their opinions on the usefulness and possibilities of state interventions. A state-internal actor cited politicization as a road- block: “We all know the Democrat[ic] Party gets a lot of money from teachers’ unions. And teachers’ unions don’t like takeovers.” Despite objections, he believed school takeover was working because otherwise you were just basically assigning lots of kids, thousands literally, to schools where basically they were pretty much ensur[ing] they never have a future. That’s kind of a hard thing to swallow, when you look at it in its stark terms. It was time to act, and the people involved had the courage to do it and we did it. 8 IJEPL11(9) 2016 In Ohio, a state-internal actor indicated a desire for school takeover, but felt it Thier, Smith, Pitts, would not be possible due to a lack of departmental resources: & Anderson Let’s be real here. We’re not ever going to … . We don’t have enough Perceptions power nor enough people or enough expertise to take 613 school of Education Governance districts and try to make them effective. It’s an impossibility, so all you can do is say, “Here’s where you are. If you’re happy with where you are, then I’m happy. … If you’re not, then you need to go to your local school board and demand some changes to make it better.” Again, this is strictly a local problem, a local hurdle, a local issue. A state-external player in Ohio painted a different landscape of how the state might intervene with chronically failing schools. He noted the Academic Distress Commissions, five-member bodies that operate in ways “similar to what you may find in No Child Left Behind, but probably kicks in a little bit more quickly.” Though local bargaining would be required to, for example, lengthen the school day, the in- terviewee described the Commissions as having “veto rights even over a locally elected board of education.” Such a characterization countered the opinions ex- pressed by many actors in locally controlled Ohio. Budget hindrances Commentary about budget hindrances supported our decision to examine quotations by sphere of influence for salience and emphasis. This topic proved salient for all three spheres. Local leaders and state-external players emphasized howpolicies should be implemented to avoid or to minimize budget difficulties. Among local and state- external spheres, taken together, 31 of 46 quotations that discussed budget hindrances addressed processes (67%), whereas state-internal actors seemed as interested in dis- cussing whatpolicies were being implemented as they were in discussinghow. All four local Tennessee leaders in our sample identified processes pertaining to their state’s mandatory funding formula as a persistent challenge. One local leader called it the “bane of every educator’s existence” and as being so “complicated and convoluted” that even its authors could not “adequately explain how it works.” Two other local leaders in Tennessee ascribed inequities to the formula, which residents— particularly those in populous areas that can consolidate tax revenues to supplement funding—accept. One actor described the formula as creating “a land of the haves and have-nots,” where teachers “in very similar schools” 30 miles apart have $15,000 salary differences. Another Tennessee actor described “political theatre” in which school leaders must defend budgets to local mayors who “nitpick” so they can cut the program deemed “most offensive to the constituency.” The actor characterized the process as an annual ritual on reducing the confidence of the public in the edu- cation system. To me it’s backwards, having grown up in a place where, as most places, ... school boards have taxing authority. It’s backwards from going out and making the case for greater revenue on a public basis and building the case for public education in the general community. This way, the district is almost on the defensive 9 for anything that any individual member of the city council wants IJEPL11(9) 2016 to make an issue out of … . The lingering doubt about how the Thier, Smith, Pitts, & Anderson schools are managing their funds kind of takes an annual toll. Perceptions A local leader in Ohio described a related challenge as “vicious” and threatening of Education schools’ abilities to budget effectively: Governance Each and every city, village, township is scrambling to put levies on the ballot to maintain police and fire and other public services … everyone’s cannibalizing one another. And the school levies, which are funded by property taxes in our state, are at the bottom of the food chain. In Oregon, local leaders faulted budget constraints for instability. One superin- tendent noted a “dramatic percentage” of annual layoffs, which disproportionately af- fects early career teachers and thwarts recruiting teachers and leaders of color. Oregon lacks a statewide system to project budgets of five years or longer. Therefore, Oregon superintendents cannot project future vacancies. As a result of annual budgeting in- stability, Oregon schools struggle to attract staff, especially from outside the state, which Oregon superintendents depend upon to diversify teaching and administrative pools. Another Oregon superintendent said the budget forced him to cut 17 days in the academic year 2013–2014, an action he called “pathetic,” because his calendar went “down to the minute” in terms of complying with state minimum seat times. In Ohio, a local leader addressed a system change in which new state testing mandates demand a shift from pencil-and-paper exams to an online platform. Once a $41 million state expense, the mandate transferred costs to districts, which now must upgrade computers, software, servers, routers, and other infrastructure for test- ing by the Partnership for Assessment of Readiness for College and Careers (PARCC). When including investments for staff professional development to administer and prepare students for these new assessments, the interviewee estimated $10 million in testing-focused expenses for the academic year 2013–2014 alone. Both state-external players in Tennessee and California converged with local leaders by focusing on processes; however, state-external players spoke consistently about local capacity gaps in effective budgeting. A state-external player in Tennessee cited a lack of evidence-based justification for investments that led to a “fair amount of funding wasted just because of lack of capacity in budgeting processes and deciding what is really going to get the most bang for your buck.” Relatedly, California enacted its Local Control Accountability Plans and Local Control Funding Formulas to build local capacity. Still, one state-external player called the new legislation “the biggest lost opportunity” because he sees the state as “washing its hands on implementation, undermin[ing] the model itself. The assumption [is] that it will all get worked out at the local level. It could, but what you’ll get is massive political conflict.” One outlier state-internal actor in Ohio said under- performing schools’ mismanagement of federal funds “made me want to cry …. They didn’t know how to use it. It’s a real capacity issue.” An external view State-external players included heads of administrators’ and teachers’ unions, a chief executive of a regional advocacy group, members from statewide councils on busi- 10