The Education Gadfly Weekly

Byron Auguste, Paul Kihn, and Matt Miller, Closing the Talent Gap: Attracting and Retaining Top-Third Graduates to Careers in Teaching, (McKinsey and Co., September 2010).

This thoughtful McKinsey report examines the advantages and feasibility of boosting the quality of the American teacher workforce by attracting more of it from the top third of the college class. Today, the authors estimate, we draw 23 percent of new teachers from this upper-tier—and in high-poverty schools, just 14 percent. After the requisite extolling of the benefits of having smarter and better educated teachers (especially for needy kids) and reminding us that high-achieving countries do far better than the U.S. at making the teaching occupation appealing to high-ability people, the authors break into the meat of their report. Here, they offer a number of strategies for upping the reputation of U.S. teachers. Some are costly (e.g., boosting salaries overall); others are less expensive in dollar terms but challenging in other ways. (For example, they suggest making high-need schools safer and better led, giving performance bonuses to top-achieving instructors, and focusing on “turnaround” and/or STEM schools.) Cautiously, the paper points to possible offsetting savings, such as targeting a larger share of the school dollar on instruction; they estimate that we could redirect $50 billion by simply lowering our high “non-educator expenditures”(e.g. admin, transport, and ancillary services) to the OECD average. They are even so bold as to suggest larger classes and more extensive use of instructional technology, which would of course mean fewer—but better paid and presumably abler—teachers. This is the kind of fresh thinking that American education sorely needs. If we were more willing to engage in it and then act on the basis of it, perhaps we’d spend less time and energy waiting for Superman.

Cheryl Almeida, Cecilia Le, Adria Steinberg, Roy Cervantes
Jobs for the Future
September 2010 

This report from Jobs for the Future analyzed all 50 states’ and the District of Columbia’s policies that guide their overall approach to and operation of alternative education programs. The authors, through a review of state policies and legislation, examined the extent to which each state’s alternative education policies incorporated seven elements that comprise what JFS deems a model alternative education program. States should:

1. Broaden eligibility to reach beyond the traditional “at risk’ student.
2. Clarify state and district rules and responsibilities to establish quality standards for the operation and management of such programs.
3. Establish a separate accountability system that holds alternative programs to common state standards and takes into account circumstances unique to alternative education.
4. Increase support for innovation by allowing alternative educations models to replicate throughout the state.
5. Ensure high-quality among staff by creating incentives for such professionals to work in alternative programs and requiring ongoing professional development.
6. Establish outside community partnerships to provide a range of academic and support services to alternative students.
7. Create a funding formula that provides alternative education programs with a greater amount of funding than received by traditional programs through state and district per-pupil payout.

What the report found was not surprising: overall states have a lot of work left to do when it comes to educating students through alternative programs.  The District of Columbia and 40 states have implemented at least one of the policy elements but no state has adopted all seven.  Ohio, along with thirty other states and DC, has put into place policies that broaden student eligibility for alternative education beyond troubled youth. No state has adopted sufficient policies to ensure high-quality staff within alternative education programs

On the seven indicators Ohio finds itself in the middle of the pack: it only met one policy element fully and another four partially (clarify responsibilities, strengthen accountability, ensure high quality staff, and enrich funding).

Overall, the report shows that while states are starting to embrace alternative education there are still fundamental policy changes that must occur in order to capitalize on all that alternative education can offer.  To read the complete report click here.

Jonah E. Rockoff & Benjamin B. Lockwood
Columbia Business School
Fall 2010

Middle schools aren’t working. At least, that’s the conclusion made by Jonah E. Rockoff and Benjamin B. Lockwood in their new study “Stuck in the Middle,” featured in the latest edition of Education Next. The pair tracked data from approximately 200,000 of New York City’s middle schoolers on their journey from grade three through grade eight during the 1998-99 through 2007-08 school years. They found that both mathematics and English language arts test scores of students who had attended K-5 or K-6 schools, then went on to attend a middle school, dropped significantly in both English and math in the students’ first year of middle school compared to their peers who attended K-8 schools. Their scores continued to drop at least through grade eight, the highest grade level the study covered, although at the significantly lower rate per year.

The researchers suggest two reasons for this disparity. First, cohort sizes (the number of students in a given grade level at a particular school; note – this is not related to class sizes) in middle schools were more than double those in K-8 schools, and the researchers hypothesize that large cohort sizes may therefore be detrimental to student achievement. Further, surveys of parents and students at the schools examined in the study indicate that both the students at K-8 schools and their parents were more satisfied than their middle school counterparts with the safety, academic rigor, and educational quality provided by their respective institutions. The researchers suggest this is indicates an inherent flaw in the middle schools studied, but they speculate little about the reasons for the existence of this flaw.

These findings – though based on achievement data of New York City students – have parallels in other places. For example, Columbus City Schools was frustrated last year with the performance of its own middle schools, as more than 70 percent of them received a D or an F on the 2008-09 report card. This study confirms that K-8 buildings – for whatever reason – may serve urban students better than middle schools. The Cincinnati and Dayton city school systems have implemented the K-8 model over the last several years. Perhaps Columbus would do well to consider similar changes to its model. Read about Rockoff’s and Lockwood’s findings here.

Each year Fordham analyzes  performance data of schools and districts in Ohio’s Big 8 cities, and provides a ranking of each city’s schools by Performance Index (PI) score, a weighted average of proficiency results among all tested students in that school. A school’s PI score gives an overall indication of how well its students are doing on the state’s tests, across all tested grades and subjects.  The PI score scale runs from 0 to 120 – the higher the score, the better achieving the school. A previous Ohio Education Gadfly analysis lifted up 16 high-performing district and charter schools (serving any grades), and today we bring you a list of the top-performing public middle schools in Ohio’s urban areas.

This ranking of the top 20 is all the more impressive considering that middle school-aged students are traditionally among the hardest to serve well (see this week’s review on “Stuck in the Middle”). Last year following the state’s report card release, the Columbus Dispatch ran an article about the district facing up to “ugly truths,” namely that none of its middle schools met federal targets in reading and math that year, that two in five (38 percent) didn’t teach a full year’s worth of material to students, and that nearly three in every four were rated D or F.

Other cities struggle as well (to varying degrees) to improve the achievement of middle school students. But there are several scattered across the state that are doing a phenomenal job. The table below lists them in order of PI score and includes the percent of students eligible for free and reduced-price lunch.

Table 1: Top 15 highest-performing urban middle schools

Source: Ohio’s interactive Local Report Card

A few things jump out from the table. First, among the top 15 are middle schools in nearly every city: Columbus, Cleveland, Akron, Toledo, Youngstown, and Canton. (Dayton and Cincinnati schools break down grades differently and have primarily K-8 buildings and high schools, rather than true middle schools.) In nearly every urban area, there exists a model (or several) when it comes to reaching middle school-aged students. 

Second, the list reflects a mix of charter and district middle schools. But it’s worth noting that given the relatively small number of middle school students that charters serve (compared to their home districts) they are overrepresented on this list. Charter schools’ freedom to set their own curriculum, school calendar and school day schedule, and behavioral and academic expectations for students undoubtedly plays an important role in their success.

Third, we’d be remiss not to note that several of the highest performing public middle schools in Ohio’s urban centers serve high percentages of poor students (as measured by free and reduced-price lunch eligibility). Ohio lawmakers, leaders, and educators should take pride in these high-performing urban public middle schools and do everything in their power to ensure conditions for their success.

You can also view a PDF of the full rankings, which includes all Ohio public urban middle schools.

Our recent study, Needles in a Haystack: Lessons from high-performing, high-need urban schools, lifted up the successes of, and tried to extrapolate lessons from, urban schools that serve large numbers of poor kids well.  But poverty exists beyond city borders. Ohio’s highest unemployment rates exist in rural communities, and Appalachian southeastern Ohio in particular has long struggled with chronic poverty.  Still, just as we find in the big cities, schools in rural areas and small towns are also succeeding at delivering large numbers of poor students to high levels of achievement.

Using data from the Ohio Department of Education, we examined the 2009-10 academic performance of the 542 public schools where 75 percent or more of the students were eligible for free- or reduced-price lunch last year. This analysis includes rural, urban, and suburban schools, and district and charter schools alike.

Chart 1 shows the distribution of students in these schools by building rating.  Far too many -- nearly half -- of these students attend a school rated D or F by the state.  But it’s promising that fully 20 percent are in an A or B school.

Chart 1: Students in high-poverty schools by building rating, 2009-10

Source: Ohio’s interactive Local Report Card

In terms of raw achievement, several of these schools are doing very well.  A school’s “Performance Index score” gives an overall indication of how well its students are doing on the state’s tests, across all tested grades and subjects.  The PI score scale runs from 0 to 120 – the higher the score, the better achieving the school.  Of the high-poverty schools we examined, eight met the state’s goal of 100 or better for their PI score.  Another 72 had a PI score above 90, which is still outstanding, especially for schools serving highly challenged populations.

Incorporating the state’s value-added measure of student progress lets us see which schools are also helping their students make adequate annual academic progress.  As chart 2 shows, 65 percent of students in these high-poverty schools attend schools that are meeting or exceeding the state’s value-added expectations.

Chart 2: Students in high-poverty schools by building value-added rating, 2009-10

Source: Ohio’s interactive Local Report Card

Combining both achievement and progress allows us to ascertain the state’s very best high-poverty schools. Table 1 lists the top 15 high-performing, high-poverty schools last school year.

Table 1: Top 15 high-performing, high-poverty schools, 2009-10

Source: Ohio’s interactive Local Report Card

Includes only schools for which a Performance Index Score and value-added data were available.

Ohio’s economy is recovering, but slowly. The issue of how best to educate children in poverty will be a front burner issue for many of the state’s public schools for years to come.  It’s encouraging to see high-poverty schools doing well, proving that they can beat the odds and deliver challenged populations to high levels of academic achievement.

The Los Angeles Times brought the concept of value-added education data front and center before the public when it conducted an analysis of teachers’ value-added scores in that city, and then published the findings – complete with teachers’ names attached – on its website.

The Los Angeles story is the most prominent example of a national movement to link student data to individual teacher performance. Federal Race to the Top dollars will fund more than a dozen states’ – including Ohio’s – efforts to incorporate value-added data into teacher evaluations. The Bill and Melinda Gates Foundation and other philanthropies, are investing millions in teacher effectiveness initiatives.  In Ohio, public value-added results are available by district, school, grade level, and subject – teachers and administrators have access to more robust student-level data. And districts participating in the federally funded Teacher Incentive Fund program will use value-added data along with other measures to award performance bonuses to teachers.

Three state education experts interviewed by The Gadfly believe it is time to use value-added here not only for informing teachers about their students’ progress but also to evaluate the teachers themselves -- to help them improve, but, if necessary, dismiss ineffective ones.

“Clearly other districts use value-added as part of evaluation of performance, but the difference between [what happened in] Los Angeles and [what takes place in a district like] Winston Salem, N.C., is that they use value-added to assess but the data are not made public,” said Thomas J. Lasley, retired dean of the School of Education and Allied Professions at the University of Dayton. “Principals have access to it but parents do not. It’s more logical to have building leaders who understand the nuances...so they can know how to work with teachers.”

 In Winston Salem, Lasley said, if low value-added student scores indicate a teacher is having problems, the teacher is placed on a professional development plan and if there’s no improvement, eventually, there’s a move toward dismissal.

While Lasley thinks it could be 20 years before value-added teacher data are accurate enough for public release, Indiana Superintendent of Public Instruction Tony Bennett plans to make the teacher data public – names and all – on the Indiana Department of Education’s website this year or next. Bennett thinks putting heat on teachers will lead to better teaching. “(If I’m a parent) I can go to the principal and say I want the best teacher for my child,” he told education policymakers in Cleveland recently.

Jim Mahoney, executive director of Battelle for Kids, the organization that pioneered the use of value-added data in Ohio, said the state’s value-added formula could be used right now for classroom-level data and evaluation. In fact, 45 districts, including Columbus, do so, he said.

But Mahoney cautions that using value-added data for evaluation purposes won’t work unless curricula and assessments are aligned, and that remains a major problem. Also, as powerful as value-added analysis can be, he said it should never be the sole measure of teacher effectiveness. Multiple measures such as attendance, student surveys of teacher performance, and knowledge of content are also important. “A lot of this is in the power of the relationship between the teacher and the student,” he said.

Beyond the classroom, Mahoney said value-added data could be used to track graduates of teacher colleges (Ohio promised in its winning Race to the Top application to connect student achievement data back to the colleges that prepared teachers) and to help clarify the ongoing debate about the effectiveness of alternatively trained teachers (like Teach For America corps members) versus those that are trained in colleges of education.

Identifying good teachers and understanding why they’re good is vital. “Some people are generating huge gains and it is not random luck,” he said. Even so, Mahoney issued a caution, “This is a very powerful measure but there’s no one measure that should be used or can capture the essence of great teaching.”

One problem with Ohio’s value-added measure, especially if it were used to evaluate and publically identify teachers, is that it is not transparent. The model in use is proprietary, and it’s not clear how the calculations are actually done.

But there are other models. Indiana uses one developed in Colorado. And Ohio’s participation in the Common Core State Standards Initiative could bode well for the future of value-added here. “As you get common standards those commons standards are going to drive us toward more common assessments and that will drive us toward more common value-added procedures,” Lasley said.

Another problem is that value-added measures are not applied to all school subjects. Right now, value-added analysis is done only in the fourth through eighth grades in reading and math, though Governor Strickland’s 2009 education reform legislation and the state’s Race to the Top grant promise to expand value-added across more grades and subjects and tie the results to the classroom level.

That’s a good thing, according to Deb Tully, director of professional issues for the Ohio Federation of Teachers. “It’s important to weed out bad teachers, she said. “[As a teacher,] I don’t want the kids the next year after they’ve had somebody terrible,” said Tulley.

Tulley said there needs to be agreement in a school district concerning how value-added data will be used. “When you locally develop a (teacher evaluation plan) everyone can agree on we don’t have a problem using value-added as one of the measures,” she said.

However, the measure must be fine-tuned first, she said, since Ohio’s current value-added model was never designed for teacher evaluation or (determining) compensation. “We are concerned about is using it incorrectly or using data that doesn’t give a clear picture of what a teacher is doing in the classroom,” she said.

If you’re clamoring to know what Americans think about myriad K-12 education issues, then you’ve just struck gold. Three recent surveys provide a plethora of opinion data on issues ranging from charter schools and teachers unions, to taxpayer-funded increases in education spending and hot-button issues like teacher evaluations.

That the surveys have varying methodologies and unique questionnaires makes it all the more significant that common themes emerged – especially those that point to shifting attitudes about teacher-related reforms (i.e., evaluations, compensation, and tenure).

The first, conducted by Education Next and Harvard’s Program on Education Policy and Governance (PEPG)—and based on information from a random and nationally representative sample of 2,776 respondents— casts the results in a political light. Researchers found that when it comes to education policy, “divisions between ordinary Democrats and Republicans… are quite minor.” Most interesting findings include that: the public doesn’t want to increase local taxes to foot the education bill (only 29 percent favored it); 62 percent “completely” or “somewhat” support retaining NCLB’s testing requirements (against just 12  percent who don’t); a majority of respondents didn’t have an opinion as to whether Race to the Top was a federal “intrusion”; 45 percent believe ineffective teachers should be fired rather than counseled; and 49 percent favored basing teachers’ salaries on students’ academic growth (compared to 26 who opposed).

The second, the well-known annual Phi Delta Kappa/Gallup Poll, was derived from telephone interviews with 1,008 adults recruited randomly. The poll shows that a majority of respondents think that improving the quality of teaching is the best lever to improve education overall. Parallel to what Education Next found, Americans rated their own local schools more favorably than schools generally. When it comes to teacher pay, 71 percent think teachers should be paid on the “basis of their work” rather than on a standard scale (a ten percent increase since 1983).

Time Magazine joins the survey action with this opinion sample of 1,000 adults. It found that 56 percent of respondents would be willing to pay higher taxes to improve schools (yes, but is that local dollars?); a sweeping majority (67 percent) think America’s schools are “in crisis”; 64 percent support student growth being configured into teacher evaluations; more than twice as many oppose tenure than support it (66 to 28 percent); and a full half believe that unions are “obstacles” to school improvement.

It’s encouraging to see several themes emerge from the data – in particular, broad recognition that teachers are paramount and that it may be time to rethink policies related to how we evaluate, pay, and retain them. Overall, the surveys indicate that the American public is willing to think fairly innovatively in this realm. Now we just need lawmakers, politicians, and teachers unions in Ohio to agree.

Last week, Policy Matters Ohio released a report on charter school accountability.  The primary finding was that when charter schools are operated by management organizations, for-profit and non-profit alike, too often the management organizations are running the show, not the independent boards that are legally the schools’ owners. 

We at Fordham will be the first to admit that Ohio charter school law allows for blurred lines of responsibility among operators, authorizers, and school boards.  In fact, Fordham testified to the Ohio House and State Board of Education last spring in support of efforts to clarify the roles and responsibilities of sponsors (aka authorizers), governing boards, and operators. 

Some of Policy Matters’ findings are, without question, worth taking seriously. For example, if some charter schools’ governing board structures are out of compliance with state law, as the report alleges, that’s a problem that absolutely needs to be addressed post haste. 

However, other items in the report raise an eyebrow about the validity of some conclusions.

First, the methodology is messy. The report compares Ohio law regarding multiple facets of the charter school program to the National Association of Charter School Authorizers' principles for authorizers. Not Ohio charter laws to the National Alliance for Public Charter Schools’ model charter law. Not Ohio laws specifically regulating authorizing to the NACSA principles for authorizers. The report is essentially a comparison of management company contracts to NACSA's principles for authorizers – apples to oranges to say the least, and more than a bit misleading.

Then there is the research, which from Fordham’s vantage point as a charter school sponsor, is flawed.  Two Fordham-authorized schools, both operated by EdisonLearning, are included in the report. Policy Matters makes three faulty claims against these schools:

Faulty Claim 1: Members of the school’s board were picked by the school operator Edison to serve as the school’s governing body (p12).

Fact:  Dayton community and business leaders banded together more than a decade ago to recruit a high-quality charter school model to the city to provide better school options for children in long-suffering Dayton Public Schools. This, only after the local teachers union vetoed the efforts of district leadership to convert five of its lowest performing charters and have Edison run them for the school district on a contractual basis. (This whole sorry episode is chronicled in our recent book Ohio’s Education Reform Challenges.)

Faulty Claim 2: The governing board relies on legal services from Edison (p14).

Fact: The school’s governing board uses the Dayton law firm Coolidge Wall for all board related matters. Edison relies on its counsel for its operational matters (and Fordham, as sponsor, relies on lawyers from Porter, Wright, Morris and Arthur). The sustainability of these legal costs is an issue all its own; especially in light of likely state spending cuts to education in 2011.

Faulty Claim 3: The school’s governing board does not insist on independent accounting of state dollars (p14).

Fact: The school’s governing board has an independent treasurer and she provides independent accounting of state dollars at each and every board meeting. All board meetings are open to the public. Both schools are audited annually by the state auditor and the final audit report is shared with Edison, representatives of the schools’ governing board and with Fordham, as sponsor.

Second, Policy Matters’ bias toward traditional district schools and away from charter schools rears its head in several places. For one, the report goes after charter school board members who are compensated for their work, but is silent on the fact that district school board members are similarly compensated; as are many other non-profit board members. The report is also quick to dismiss the positive academic performance of more than a few charter schools, saying that just “a handful of charters have solid academic records.” In fact, in Ohio’s urban areas, where most charter schools are located, charters often perform on par with or better than their district peers. The Dayton Daily News, for example, ran a story just this month under the headline “Eight of 10 top public schools in Dayton are charters.”

The anti-charter bias reemerges in the author’s description of Fordham as "a national think tank with a free market orientation toward education reform." Yes, Fordham is a staunch supporter of school choice, but we’ve been saying for years (in fact, gasp, for a decade) that choice for choice’s sake isn’t sufficient. There must be high standards and accountability for all public schools, and persistent failure isn’t an option. In fact, we take this so seriously that, as an authorizer, we have closed three schools and put several others on notice.

It’s encouraging when a left-of-center organization like Policy Matters joins our camp and call for smart improvements to Ohio’s charter school program (as it does in this report), rather than the obliteration of charters altogether.  But this report has so many faults in terms of its methodology and facts that it hurts the cause of a sensible middle ground by giving ammunition to both ardent anti-charter foes and zealous supporters of all charters irrespective of quality or performance.

Cheryl Almeida, Cecilia Le, Adria Steinberg, Roy Cervantes
Jobs for the Future
September 2010 

This report from Jobs for the Future analyzed all 50 states’ and the District of Columbia’s policies that guide their overall approach to and operation of alternative education programs. The authors, through a review of state policies and legislation, examined the extent to which each state’s alternative education policies incorporated seven elements that comprise what JFS deems a model alternative education program. States should:

1. Broaden eligibility to reach beyond the traditional “at risk’ student.
2. Clarify state and district rules and responsibilities to establish quality standards for the operation and management of such programs.
3. Establish a separate accountability system that holds alternative programs to common state standards and takes into account circumstances unique to alternative education.
4. Increase support for innovation by allowing alternative educations models to replicate throughout the state.
5. Ensure high-quality among staff by creating incentives for such professionals to work in alternative programs and requiring ongoing professional development.
6. Establish outside community partnerships to provide a range of academic and support services to alternative students.
7. Create a funding formula that provides alternative education programs with a greater amount of funding than received by traditional programs through state and district per-pupil payout.

What the report found was not surprising: overall states have a lot of work left to do when it comes to educating students through alternative programs.  The District of Columbia and 40 states have implemented at least one of the policy elements but no state has adopted all seven.  Ohio, along with thirty other states and DC, has put into place policies that broaden student eligibility for alternative education beyond troubled youth. No state has adopted sufficient policies to ensure high-quality staff within alternative education programs

On the seven indicators Ohio finds itself in the middle of the pack: it only met one policy element fully and another four partially (clarify responsibilities, strengthen accountability, ensure high quality staff, and enrich funding).

Overall, the report shows that while states are starting to embrace alternative education there are still fundamental policy changes that must occur in order to capitalize on all that alternative education can offer.  To read the complete report click here.

Jonah E. Rockoff & Benjamin B. Lockwood
Columbia Business School
Fall 2010

Middle schools aren’t working. At least, that’s the conclusion made by Jonah E. Rockoff and Benjamin B. Lockwood in their new study “Stuck in the Middle,” featured in the latest edition of Education Next. The pair tracked data from approximately 200,000 of New York City’s middle schoolers on their journey from grade three through grade eight during the 1998-99 through 2007-08 school years. They found that both mathematics and English language arts test scores of students who had attended K-5 or K-6 schools, then went on to attend a middle school, dropped significantly in both English and math in the students’ first year of middle school compared to their peers who attended K-8 schools. Their scores continued to drop at least through grade eight, the highest grade level the study covered, although at the significantly lower rate per year.

The researchers suggest two reasons for this disparity. First, cohort sizes (the number of students in a given grade level at a particular school; note – this is not related to class sizes) in middle schools were more than double those in K-8 schools, and the researchers hypothesize that large cohort sizes may therefore be detrimental to student achievement. Further, surveys of parents and students at the schools examined in the study indicate that both the students at K-8 schools and their parents were more satisfied than their middle school counterparts with the safety, academic rigor, and educational quality provided by their respective institutions. The researchers suggest this is indicates an inherent flaw in the middle schools studied, but they speculate little about the reasons for the existence of this flaw.

These findings – though based on achievement data of New York City students – have parallels in other places. For example, Columbus City Schools was frustrated last year with the performance of its own middle schools, as more than 70 percent of them received a D or an F on the 2008-09 report card. This study confirms that K-8 buildings – for whatever reason – may serve urban students better than middle schools. The Cincinnati and Dayton city school systems have implemented the K-8 model over the last several years. Perhaps Columbus would do well to consider similar changes to its model. Read about Rockoff’s and Lockwood’s findings here.

Byron Auguste, Paul Kihn, and Matt Miller, Closing the Talent Gap: Attracting and Retaining Top-Third Graduates to Careers in Teaching, (McKinsey and Co., September 2010).

This thoughtful McKinsey report examines the advantages and feasibility of boosting the quality of the American teacher workforce by attracting more of it from the top third of the college class. Today, the authors estimate, we draw 23 percent of new teachers from this upper-tier—and in high-poverty schools, just 14 percent. After the requisite extolling of the benefits of having smarter and better educated teachers (especially for needy kids) and reminding us that high-achieving countries do far better than the U.S. at making the teaching occupation appealing to high-ability people, the authors break into the meat of their report. Here, they offer a number of strategies for upping the reputation of U.S. teachers. Some are costly (e.g., boosting salaries overall); others are less expensive in dollar terms but challenging in other ways. (For example, they suggest making high-need schools safer and better led, giving performance bonuses to top-achieving instructors, and focusing on “turnaround” and/or STEM schools.) Cautiously, the paper points to possible offsetting savings, such as targeting a larger share of the school dollar on instruction; they estimate that we could redirect $50 billion by simply lowering our high “non-educator expenditures”(e.g. admin, transport, and ancillary services) to the OECD average. They are even so bold as to suggest larger classes and more extensive use of instructional technology, which would of course mean fewer—but better paid and presumably abler—teachers. This is the kind of fresh thinking that American education sorely needs. If we were more willing to engage in it and then act on the basis of it, perhaps we’d spend less time and energy waiting for Superman.