The Modern Cult of Efficiency

The Historical Roots of Educational Intermediary Organizations

In 1899, the Bethlehem Steel Company contracted with one such external expert, Frederick Winslow Taylor, a mechanical engineer who was hired to assess the factory’s operations and make recommendations for minimizing inefficiencies through better measurement, surveillance, and presumably scientific, data-based procedures (Callahan, 1962). Taylor’s approach, later dubbed “scientific management,” was characterized largely by methods for distilling work into discrete, quantifiable tasks; measuring observable outputs; exercising heavy managerial control over workers; and minimizing costs by appealing to workers’ economic self-interests, as well as by engaging in systematically derived best practices and planning. His system was swiftly taken up by business and, shortly thereafter, the social sectors.

Several conditions coalesced at that time to create a national context that was ripe for scientific management in industry, education, and beyond: a booming capitalistic industrial revolution, fueled by Americans’ dominant economic philosophy of free enterprise; a mounting public intolerance for stories of their industrial titans’ waste and corruption, dug up by muckraking journalists; and a growing concern over how to design America’s system of schooling for a diverse society undergoing an influx of immigration (Callahan, 1962; Tyack & Cuban, 1995). Together, the developments set the stage for “reformers” to demand more transparency in business, political, and social leaders’ affairs, as well as a greater reliance on business methods and principles to solve inefficiencies and maximize productivity in order to better serve business and the nation’s economic interests. Educational administrators found themselves stuck squarely in the middle of this reform movement.

By the turn of the century, popular magazines, newspapers, and business leaders regularly implored educational administrators to avail themselves of lessons from big business (Callahan, 1962; Cuban, 1988). Reformers advocated for the construction of quantitative metrics to measure schools’ “products” and to employ economic logic to guide the educational enterprise. Monitoring, testing, and competition soon permeated public education, and the practice of hiring “efficiency experts” to collect data on schools’ operations, evaluate performance, and make recommendations to maximize productivity became commonplace. From that point on, the bonds between external consultants from business and industry and school administrators were fixed. No longer were educators and communities left to their devices to deliberate about and solve their own problems; individual consultants or firms were regularly hired to collect data from the schools, pinpoint the errors in their ways to their leaders, and design reforms intended to tighten up the bureaucratic slack (Cuban, 1988). Urban schools were particularly susceptible to these managerial reforms, for it was in these settings—usually occupied by large numbers of immigrants, non-native speakers, and children from low-income communities—where performance was deemed to be lacking and where the media and politicians diagnosed a need for better management.

The Contemporary High-Stakes Accountability Policy Context

More than a century later, a sizable body of evidence details the failures of these reforms to achieve their stated goals (Tyack, 1974; Tyack & Cuban, 1995). Yet many of today’s self-styled reformers continue to closely resemble these early efficiency experts. Across the country, educational intermediaries are often hired to broker change between schools and districts (Trujillo & Woulfin, in press). Intermediary organizations “occupy the space in between at least two other parties” and “primarily function to mediate or to manage change in both those parties” (Honig, 2004, p. 67). While intermediaries include both non-profit and for-profit providers of technical assistance and other educational resources, in this article, “intermediary” refers only to non-profit organizations that provide coaching, consulting, or professional development, and that do so by brokering these resources between schools and district central offices.

Non-profit intermediaries have proliferated in recent years. In fact, an entire school improvement industry has grown alongside the public school system with the primary purpose of strengthening educational performance using methods and resources that, presumably, the system lacks (Rowan, 2002). In California, this industry is thriving, and intermediaries are increasingly assuming responsibility for brokering managerial expertise, usually in the most struggling schools (Trujillo & Woulfin, in press).

Intermediaries usually refer to themselves as “school reform” organizations or agencies. These days they often align their assistance and support with federal education requirements—high-stakes accountability policies grounded in the principles of efficiency, productivity, and accountability (Burch, 2009). In fact, these policies have solidified intermediaries’ roles in public education by writing in specific mandates for districts and schools to hire the agencies as a condition of their compliance with high-stakes accountability regulations (Burch, 2009; Lipman, 2004). Consistent with the cornerstones of the No Child Left Behind Act and, more recently, Race to the Top programs, today’s intermediaries fill a policy-driven niche: They continue to serve as external experts who promote schools’ use of measurable outcomes, standardized processes, and observable, quantifiable indicators of test-based effectiveness (Trujillo & Woulfin, in press).

These national policy structures notwithstanding, the nearly universal trends among states toward standards-based accountability—systems of standardized content, assessment, target-setting, and sanctions for low test performance—have cultivated local policy contexts that are conducive to intermediaries’ interventions. Because districts and schools that serve historically low-scoring students—poor children and children of color, English Learners, and immigrants—experience the greatest pressures to step up test results and other monitored outcomes, many contract with intermediary organizations to meet performance demands and guard against the policies’ punitive consequences (Burch, 2009; Renée & Trujillo, 2014).

Data as Reform: The Primacy of Checklists, Rubrics, and Other Pretty Papers

Based on the literature on New Public Management, namely, the research showing how the new managerialism focuses individuals’ sights on regularly collecting and analyzing quantitative data to measure, monitor, plan, and evaluate performance, I examined the intermediary’s overall aims for its schools, as well as the day-to-day activities it used to reach those aims. When asked to articulate their goals, intermediary staff consistently responded that they aimed to increase educational equity in low-income public schools that serve primarily students of color and English Learners. In their words, they supported schools to deliver rigorous, culturally relevant, standards-based curriculum for these students by using practices that are responsive to their academic and emotional needs. However, when asked to describe the nature of their day-to-day services, both the agency and school personnel almost universally cited technical, managerial activities divorced from questions of rigor or pedagogy that was culturally relevant or specific to English Learners. Their reforms were circumscribed around discrete tasks for target-setting, analysis of standardized test data, standards-driven curricular planning, and the establishment of performance monitoring systems for both principals and intermediary staff. Its approach was embodied by the artifacts of its reforms: spreadsheets, checklists, matrices, diagrams, rubrics, and other graphical displays of data and content standards were by far the most frequently cited examples of the services they introduced to schools.

From the intermediary’s point-of-view, this managerial focus was purposeful. It equipped teachers and principals with general tools that were necessary to carry out their work. This staff member put it this way,

We don’t do content work. We don’t teach people . . . we don’t do literacy instruction. We put structures and processes in place to assist teachers to be able to better do their jobs around collaboration and results and data.

Another participant from the intermediary summarized their work similarly:

Our brand is the structures and processes that we help put in place for teachers to be able to collaborate, to examine data, to use that data, to go back and then drive their instruction and to help the principals and the leadership team do the same thing.

Other staff members reinforced this focus. They explained that because they were “lighter touch,” they usually provided schools with test score data or graphic organizers to form the basis of principal and teacher coaching, staff development, and leadership monitoring. As one participant captured the essence of their work, “We are strongest in the technical details, like how you create a monitoring system.”

The intermediary crafted these monitoring systems around instruments for data collection and analysis in schools. Rubrics, scorecards, and other data collection spreadsheets comprised the bulk of these tools. Most staff members we interviewed proudly cited one such rubric that spanned 60 pages in length. One of them explained the origins of this rubric that was used for both school leaders’ reflection and internal monitoring of the reforms’ progress:

We [created] a 60-page rubric that has three columns. It’s pretty comprehensive. We had a consultant help us define it . . . We did a bunch of external research, as well as studied our own schools. We looked at other people’s rubrics, like from Teach For America and New Leaders for New Schools, and we tried to understand what they were defining as strong instruction. We brought all that together to create this rubric.

The intermediary used this particular rubric to collect data from multiple perspectives and then crunch the numbers to arrive at a score to judge each school’s performance. “We took those sources of data and came up with one triangulated score . . . We took that score back to the schools, had conversations with the principals and said, ‘This is what the scores are showing.’” Staff regularly and enthusiastically cited this rubric as an example of their progress in becoming more data-driven in their evaluation of their schools as well as their brand of school reform.

School personnel, however, were often less sanguine about the utility of this and other rubrics that the intermediary introduced. One principal of a high-scoring school echoed others when he communicated his dissatisfaction with the quantity and quality of documents:

Organizationally they’ve been frustrating. Don’t get me wrong. I also think they’ve been helpful, but I feel like my community views them as, who are those people? Something I would change is the quantity of pretty things they make me fill out. I don’t keep any of them. I use them all as recycle paper . . . I don’t think it helps us to make another visual . . . like the one I just threw out. It was this laminated thing with concentric circles . . . throw that in the trash. There are so many things on their documents that they make you feel like a huge failure as a school. I don’t think we’re failing, but when you look at how many things they have you fill out, like where you are on the rubric, you’re like wow, we’re a two on this rubric across the board.

A different principal recalled the templates that the intermediary designed for teacher meetings as follows:

[They] created a lot of templates for our professional learning communities. During PLC time they’d say, “These are some of the outcomes and this is our process for getting there.” So they created a lot of forms. They were helpful, but honestly, sometimes it was just too much writing or they just asked 15 questions when they could have just asked five. I saw teachers look and say, “Wow, this is a lot of work . . .” They’re already really overworked and I could see the stress.

Another school leader shared that, although the intention of some of the intermediary’s activities was useful, the language used by its staff, like that of the district, was often inaccessible for teachers. She reflected,

I intentionally avoid [the intermediary’s] and district phrases with my teachers. For example, I said to them, “If you use the words theory of action, they will not know what you’re talking about, but if you ask them, are you doing X, Y and Z, or what’s the big focus of the school, they’ll know.” I don’t use the word theory of action because it daunts the teachers a little bit. And you get a little eye roll when you bring up that or their rubric.

Other principals expressed similar sentiments. In response, some confided that they tried to buffer their teachers from parts of the intermediary’s tasks by filling out checklists, rubrics, and other documents themselves so that the intermediary, like their district, would leave their teachers alone.

Despite the inconsistency between the agency’s perceptions and their schools’, staff members continued to develop tools to quantify schools’ performance to prescribe strategies for improvement. They also aimed to measure each school’s reform fidelity through similar methods, as this participant explained:

We use a database approach to determine how much of the work should be implemented in order for it to be considered high fidelity . . . Because our schools that have 75 percent implementation of the prioritized strategies got double-digit [standardized test score] growth, we set our [fidelity] target at 75. The implementation of all schools except one exceeded 75 this year.

In this case, a desired level for fidelity of implementation was defined by correlating the percentage of items in a database with schools’ gains on standardized tests. This arbitrary target—double-digit test score growth—was deemed to be a valid reflection of schools’ quality and, hence, a meaningful correlate of all schools’ implementation fidelity. One of the intermediary’s leaders elaborated on this technique:

We [measure] this twice a year, so our through-line then is that high quality implementation leads to increases in capacity at the school site, which will lead to increases in academic achievement.

This quote reveals several assumptions in the intermediary’s logic. Their leader assumed that the information stored in the database accurately reflected the reality in schools and focused on the correct indications of more equitable teaching and learning. She also assumed that the standardized test scores represented the best outcome by which to gauge success. Nevertheless, this particular tool was used not just to measure implementation fidelity, but to stipulate the next steps for intermediary staff. The leader went on:

Individual [intermediary staff members] are assigned specific actions that are referenced in the tool, and depending upon the item that’s there and the duration and the time of the year, twice a week implementation is assessed against the goals for that particular step.

Thus, implementation was monitored according to a collection of observations that were counted as often as twotimes each week.

Strategic plans for each school were distilled into matrices that specified lists of strategies to be used at each site, depending on their placement in respective cells. Most intermediary staff spoke highly of these managerial tools. They expressed pride in their construction and believed that they would help focus schools’ efforts on the most effective strategies for boosting test scores. Yet this stance may have conflicted with the intermediary’s broader goals for increasing educational equity for students of color and English Learners by promoting practices that are responsive to their academic and emotional needs. Despite these more ambitious goals, most intermediary staff appeared content and even pleased to facilitate activities for regularly collecting, organizing, and analyzing data to measure performance—hallmarks of the audit culture that Apple and Ball describe. Still, a minority of intermediary staff dissented. These staff usually communicated that the instruments were too general in nature to be tailored to schools’ unique needs. They also worried that they may be measuring the wrong indicators of school quality all together. This dissenter captured the following concerns:

It’s too general. Are practices in place that lead to greater student engagement? Are culturally relevant practices being used? Without training and without real definitions and without thought leaders who really educate everyone in the school, somebody who doesn’t understand this or who has a low skill level may interpret it very differently than somebody who does understand it.

For this staff member, the data generated from the organization’s matrices and rubrics were at risk of being misconstrued if individuals lacked the professional knowledge to accurately interpret them. She went on to lament these same tools’ lack of specialization by explaining that strategies that may apply in one context may not necessarily be the best ones for another context, although the tools themselves did not differentiate significantly across schools. In this case, an intermediary’s staff member had not yet aligned her own values and beliefs to be consistent with the logics of the neoliberal audit culture, but she was in the minority within her organization.

In these ways, the intermediary’s reforms closely embodied the principles of the new managerialism. They concentrated almost exclusively on outputs, planning, and continuous monitoring, like the literature would predict. Implicit in most of the reforms was an assumption that the data in fact yielded almost perfect information or at least information that was reliable enough to define which predetermined set of actions would be implemented across contexts. Principal and teacher concerns notwithstanding, the intermediary’s staff often boasted of their progress in refining their “brand” of reform through these practices and instruments, but the ways in which the tools led to more socially just schools was less apparent. Rather than serve as a means for creating more equitable teaching and learning, the reforms and the tools used to guide them appeared to function as ends in themselves. Goals for improving instruction for students of color and English Learners seemed to become obscured along the way.

Management as Leadership: Reductive Models for Leading and Teaching

Next, in light of the literature on the managerialist characteristics of leaders and teachers that are fostered under neoliberal policy regimes, I considered the ways in which the intermediary framed good leadership and teaching. The tools that the intermediary employed, as well as the focus of its reform activity, carried particular messages about what strong leaders and teachers look like. In fact, the intermediary staff stressed that cultivating effective leadership was at the core of their reforms. When asked how they defined a good leader, staff consistently recalled the same characteristics. This participant’s response represented many:

What is a good school leader? In our rubric we’ve defined it in basic terms. A good leader implements results-oriented . . . inquiries. They have a vision. They set goals. They lead planning. They monitor. They look at data to see how effectively the plan is going. It’s essential that a leader is doing that in terms of our program.

This quote summarizes almost every staff members’ thinking about the roles of principals in schools. Good leaders almost exclusively prioritized the technical aspects of school change, in their view, and their rubric reinforced these messages to schools. The intermediary judged principals according to these criteria, and school and district actors with whom it contracted did the same. The tool may have also captured qualities related to trust, relationships, values, or the socio-political nature of principals’ work, but neither staff nor educators ever referred to them.

As for teaching, the focus of the agency’s services implied a fairly narrow conceptualization as well. When asked to describe the nature of their interactions with the organization, most teachers recalled the same set of experiences. Forms that summarized test score data and that helped plan instruction according to the state content standards formed the core of their work with the intermediary staff. Perhaps because these teachers were all working in urban schools that were under heavy pressure to raise their test scores, several found their products at least moderately helpful. This teacher reflected on the intermediary’s role this way:

Some of the things from [the intermediary], like the curriculum [forms], we use those to prepare for state testing and they’re very helpful. They were great because they’re all standards-based. You can parse out kids on each specific standard. And those [state] tests are indicative of what [the intermediary] is all about—a very standards-based approach to instruction.

For this teacher, the organization helped him meet testing demands by distilling his curricular decisions and sorting students according to specific content standards that were to be tested. The message about the purpose of his teaching in this case was clear: to prepare students for the state assessment. His reflections also encapsulate the values and beliefs of Apple’s (2007) “new auditees.” Despite the ways in which strict standards-based curricula and testing diminished his professional judgment and autonomy as a teacher, he welcomed the intermediary’s tools for curriculum alignment, student sorting, and test preparation because these were the resources that helped him cope within a competitive education policy environment.

Staff members corroborated teachers’ perceptions of the intermediary’s focus. This employee summarized several of her colleagues when she stated, “The materials and resources that we provide are kind of packaged standards, ways of looking at standards, making things . . . teacher friendly.” One of her colleagues expanded on this when he noted, “We give [teachers] some sample standards for them to talk about. How will a child demonstrate mastery? We give graphic organizers for them to break down the standard and make sense of it.”

In this way, the organization framed teaching as a collection of decisions that existed within the confines of the state’s regulatory framework. Content standards and test scores formed the basis for much of teachers’ exchanges with the intermediary, but how they interpreted this information was left to the teachers. In fact, several intermediary staff clarified that this generic characteristic of their services was purposeful. The following leader summed up this thinking:

We don’t have a firm stance on literacy instruction, per se. We have a set of practices that can be filled out in the rubrics and that are actually fairly generic practices that apply across content areas. There are things about how you differentiate, how you engage students, how you kind of use a gradual release model. We intentionally have stayed out of what kind of reading models they should be using. We want to work with a diverse range of districts . . . [so] we have a set of practices that are fairly generic . . .

This quote not only shows how the organization supported teachers in concrete ways but also shows why it opted for more standardized practices. More specialized supports, in its view, would limit its potential to contract with more districts. Generalized products, like the ones its staff described, carried a greater potential for application in more settings. Within the Education Management Industry (Gunter, 1997), standardized “commodities,” especially those for data-based decision making, increased the intermediary’s marketability.

Another staff member commented, “We like to stay pretty neutral . . . If they want to use a curriculum like Open Court we’re going to support that. We’ll push them to use data and adapt it, though.” He went on to clarify that their most defining work entailed “bringing that standards-based lens to everything that we’re doing. We offer tools, even just agendas, and those are pretty standard across our schools.” However, their seemingly impartial tools for standards-based, test-centered instruction implicitly conveyed specific messages about which knowledge was most valuable. By guiding practice according to state-mandated content, the intermediary reduced teachers’ work to the delivery of centrally defined, not locally driven knowledge. Opportunities for teachers to debate and discuss their priorities for what students should construct, beyond test-centered knowledge, were unaddressed by their reforms. By setting the parameters of teachers’ lessons according to a mix of state mandates and an outside agency’s priorities, the organization also implied that teachers lacked the requisite expertise to collectively determine which knowledge counted as valid. In this case, as Gunter (1997) explains, a central authority determined the knowledge that was deemed valid and teaching was not so much a professional endeavor as it was a managerialist one.

While most intermediary staff interpreted this “neutrality” as an asset, here, too, a small number acknowledged the tensions in this stance with respect to the organization’s stated goals around educational equity. One of the most senior leaders put it this way:

We’re an organization that aspires to be something that at this point we’re not, which is a truly culturally reflective and culturally responsive and culturally respectful organization. We want to set ourselves apart from organizations that provide a kind of colonial approach to providing support to schools, one that by design acknowledges the assets that are in the space and grows and supports them from that perspective, but we need to build our skills to do that. Many of our [staff] are ex-TFAers, so they come with that imprimatur and don’t necessarily have a broad base of experience with different kinds of people in different contexts. But at the same time, we’re at a place of consciousness on that front and are building our knowledge and our skill.

She went on to explain that because they often hire from Teach For America (TFA), their staff is really good at teaching people to develop strong objectives and assessments and use a gradual release of responsibility model—competencies TFA develops in their recruits—but their brief experience in education limited their notions of what best practices looked like. A colleague added a similar sentiment in the following quote:

We’ve had trainings from time to time and there’s awareness of the general body of knowledge around culturally relevant teaching and some authors and researchers, but I think the organization is pretty loose. I don’t think the organization has said, “This is our stance on that and these are practices that are non-negotiable and that make culturally relevant teaching effective.” We need more thought leaders around that in the organization . . . I think a lot of our schools don’t even see us as a leader of culturally relevant teaching and as advocates for kids of color. I think they look at us and say, “You guys are smart system thinkers who can analyze data to no end.” I think they value us, but I think they really do need to see us as people who are advocates for communities of color, or who are very knowledgeable and bring a lot to the table around what it takes actually have transformational change, so we’re going to just get incremental change. That’s honestly what we have been getting the last several years—incremental change.

A small number of principals also confided about this weakness in the intermediary’s work. They attributed it, in part, to the composition of the staff. More often, however, they commented on the limitations as they applied to the work more generally, as this principal did here:

Their staff tends to be fairly young as far as experience is concerned. Whereas [the staff member assigned to my school] is really smart, she’s about 30 . . . I don’t really feel she can help me, to be totally honest, on administrative stuff. She’s a great thought partner on pedagogy for how to motivate people in professional development, but I don’t think she really understands what it’s like to run a complex school, so there’s a contextual difference there that I don’t find that helpful.

Thus, while most principals saw how the intermediary provided pockets of help to schools, several acknowledged the limitations of their services. The organization’s staff who lacked deep professional knowledge of issues around race and culture, as well as around the nuances of leading school change, were seen as restricted in their potential to help guide the most sophisticated aspects of equity-minded improvement. The result was a collection of managerial services that hovered around the surface of schools, but that were viewed by some as not reaching the most difficult dimensions of organizations, leadership, and goals for social justice.

Competing for the Market Niche: Business Logics, Roles, and Language

While the specific skills and expertise of the intermediary’s staff may have partially explained the shortcomings of their approach to schools, their reasons for selecting particular reforms and organizing their internal and external activities were, perhaps, even more revealing. When reflecting on the emphases of their strategies, all staff demonstrated clear market-oriented thinking and a keen sense of competition among intermediaries to establish themselves as leaders in a particular niche of what Gunter called the Education Management Industry. Their logic, internal forms of organization, and language all illuminated this thinking.

Staff repeatedly shared that their specialization in “equity” and focus on students of color made them stand out among peer organizations. At the same time, every staff member noted that the intermediary aimed to develop a clear, replicable, and scalable “brand” of school reform. The brand, as this article details, consisted largely of structures and systems for data-driven decision making. Some staff spoke with pride about superintendents’ use of their laminated tools. These same staff referred to the district chiefs as their “best sales people” who could most effectively spread the word about their “brand” to other districts and schools. Indeed, one of the intermediary’s goals was to scale up their services to increase their “market share.” Yet some staff identified considerable tensions in this goal. They worried that they had not yet developed sufficient expertise in their model of reform to expand it to new contexts. Their model, in their eyes, continually shifted to meet funders’ preferences, and they had not devoted sufficient time to testing their most current model to know whether it contained the right recipe of ingredients to help larger numbers of schools.

Although the intermediary’s focus may have reflected funders’ changing preferences, several staff also perceived funders to be heavily concerned with rapid test gains and recognized a related dilemma. This staff member characterized the pressure to balance deeper, long-term support with techniques that quickly boosted scores:

You feel this desperate need to get results in one year, and sometimes you take shortcuts when you really need to take a step back and help the school be more strategic, but it’s hard because we feel pressure to get results in one year . . . because all of our funding depends on it.

Layered on top of these financial pressures were scalability pressures. In response, one of the organization’s leaders took several steps to codify the intermediary’s products. She reasoned,

If there is an actual program that’s codified and . . . that’s static, that can be replicated and that is key to the industry . . . Then the throughput—that line to results—is crystal clear [because] you know what that program is really aiming for and is expected to produce.

Other staff also surmised that a standardized set of products could help clients distinguish its reform work from other reforms taking place in schools. One employee put it like this:

There’s got to be a . . . codified approach that everybody knows how to do . . . in as similar a manner as possible with different human beings. I don’t want us to look like McDonald’s arches . . . but a person should be able to walk into that school and know that there are processes and structures in place that are aligned and recognize that there’s something that’s going on there that’s different than in another school . . .

Standardized rubrics, data analysis templates, agendas, and other graphic organizers formed the bulk of this codified approach, although internal management structures were also developed to further aid such growth. Staff regularly referred to their internal performance management structure that was intended to add rigor to internal evaluation and expectations for employee performance, and that could be replicated in new settings.

Variation in staff roles and daily activities was minimized to ensure that employees who occupied the same position allocated similar percentages of time among tasks. Titles were revised to align better with private sector roles and were redistributed hierarchically. Employee evaluations were conducted more frequently, based on more forms of data collected in schools, and tied to new salary structures. The head of the intermediary self-described her role as “a leader of managers.” In conversations, she regularly stressed the importance of more clearly delineating her staff’s managerial responsibilities and empowering them to solve problems before communicating up the organization’s chain of command. Some staff expressed concern that the organization may have become too tiered and rigid in its quest to codify its program, establish its niche in the industry, and scale up its work.

These competitive concerns were evident in the language intermediary staff used to communicate its work. Terminology from the corporate lexicon was commonplace. Talk of human capital and performance management was frequent. References to throughputs, value propositions, onboarding, insourcing, market shares, results orientations, scorecards, management philosophies, coherence, alignment, delineated salary levels, performance metrics, targets, high-leverage strategies, light touch approaches, benchmarks, risks associated with investments, brands, and mathematical formulas for performance appraisals were heard repeatedly throughout interviews. Language assimilated from the policy environment was as common: performance bands, proficiency, and talk of effectiveness were nearly universal among employees.

Based on this language, school improvement appeared to be a largely market-oriented affair. Terminology specific to pedagogy, public goods, or socio-political aims for education, while present, was far less readily employed when staff reflected on the nature of their work. From these patterns, the work of the intermediary was a business, and the business of intermediaries was school reform.