College for all has been the goal of K–12 schools for at least twenty-five years. This has meant that America’s schools typically do not provide young people with work experience. This experience gap has young people leaving high school with little understanding of work and practical pathways to jobs and careers. They also find it difficult to transition from school to work and earn lower wages when they begin work.
Many Americans, including young people, no longer believe college is the default pathway to success. They want K–12 schools to provide practical learning. For many, this means apprenticeship programs, making them a new learning campus for young people.
Unlike the conventional college campus, students in apprenticeships have a full-time job; earn a living; learn from a mentor in the workplace and classroom; and receive a credential with little to no student debt. This learn-and-earn model is spawning new forms, like pre-apprenticeships, youth apprenticeships, and apprenticeship degrees.
The demand side: What Americans want
Populace reports that Americans’ current priority for K–12 schools is ensuring students develop practical, tangible skills that prepare them for a career. Only one in four (26 percent) think schools do this. Americans believe that apprenticeships develop this practical sense, with strong support from the public, including young adults, especially Gen Z.
More than nine in ten (92 percent) Americans view apprenticeships favorably, while more than six in ten (62 percent) say they make people more employable than going to college. Seven in ten (71 percent) disagree that an apprenticeship limits future employment options, with more than nine in ten agreeing (94 percent) that apprenticeships help individuals obtain new careers. When asked to choose between a full-tuition college scholarship and a three-year apprenticeship leading to a good job, nearly six in ten (56 percent) parents opt for apprenticeships. And more than two-thirds of Gen Z high schoolers say post-high school learning should be on the job through internships or apprenticeships (65 percent).
The supply side: Apprenticeship programs today
In 1937, President Franklin Roosevelt signed the National Apprenticeship Act, which gave the federal government authority to register and oversee apprenticeship programs, with states having an option to do so. About half the states exercise that option, with 27,000 registered programs enrolling around 500,000 individuals.
The U.S. lags behind many nations in using apprenticeships for workforce preparation. Moreover, around 70 percent of U.S. registered programs are in construction trades, such as carpentry and plumbing. This is unlike other English-speaking nations like Australia and the United Kingdom where most government-supported apprenticeships are in fields like healthcare, logistics, technology, and the financial sector. The U.S. also has pre-apprenticeship programs that prepare individuals for registered apprenticeships and youth apprenticeships, typically part of a young person’s high school experience (some are registered). Apprenticeships differ from internships, which typically are short-term, entry-level, often unpaid jobs without a mentor and no industry credentials.
State and federal support are growing for apprenticeships. They have bipartisan backing is states as diverse politically as California, Colorado, Tennessee, and Texas. Federal spending for apprenticeships in the Labor Department’s Employment and Training Administration more than doubled over five years, from $90 million in 2016 to $185 million in 2021.
These programs succeed in preparing individuals for rewarding employment. One study of registered apprenticeships shows that workers earn $240,000 more over their lifetime—$300,000 when including benefits—by participating. Another study documents how states are creating new pre-apprenticeship programs as short as one to three weeks or as long as eight weeks that recruit a more diverse pool of underrepresented groups for apprenticeships.
The supply side: Creating more apprenticeships
The apprenticeship intermediary is a new third-party structure at the center of U.S. efforts to expand apprenticeship programs. Ryan Craig, managing director at Achieve Partners and author of Apprenticeship Nation, describes these organizations as “earn-and-learn” intermediaries that use a “hire-train-deploy” model to run and pay the up-front costs of apprenticeship. They can be public, non-profit, or for-profit organizations—for example, community colleges, chambers of commerce, and commercial staffing companies—which work locally, regionally, statewide, or nationally. Companies would “try-before-you-buy” an apprentice, paying a fee to the intermediary for its work. In short, an intermediary is a sum of three functions: “Earn-and-Learn” + “Hire-Train-Deploy” + Try-Before-You-Buy = Apprenticeship Intermediary.
The supply side: Apprenticeship degrees
Some countries integrate degrees into apprenticeship programs. For example, the United Kingdom has an earn-and-learn apprenticeship degree that takes between three and six years to complete and leads to a debt-free bachelor’s or master’s degree in fields like health and sciences, business and administration, and aerospace. Individuals apply to an employer, with the degree granted by a partner university. The government pays two thirds the cost; the employer the other third.
U.S. K–12 education is adapting this approach to create debt-free teacher apprenticeships that award bachelor’s and master’s degrees. The nonprofit Reach University is a leader in this, with its affiliates Oxford Teacher’s College undergraduate school of education and Reach Institute graduate school. They offer eight degrees and certificate programs that enroll over 1,500 students. The federal Pell Grant typically covers all but $2,000 in costs, with philanthropy or U.S. Department of Labor funds covering most of the balance, with a student contribution capped at $900 a year. The U.S. Departments of Education and Labor are creating paid registered apprenticeship programs for teaching, with the first program approved in Tennessee.
Opportunity pluralism
Apprenticeships create opportunity pluralism, where many pathways lead to human flourishing rather than the one route of college for all. In refusing to equalize opportunity along the single route of college for all, these earn-and-learn apprenticeships make the nation’s opportunity infrastructure pluralistic.
They develop in young people two important elements for lifelong success: knowledge and relationships. As the adage goes, it is not only what you know but also who you know. Moreover, this earn-and-learn approach equips young people with not only knowledge that pays but also relationships that are priceless. The goal for K–12 schools is no longer only college for all but earn-and-learn opportunity pluralism where many pathways lead to human flourishing.
News stories featured in Gadfly Bites may require a paid subscription to read in full. Just sayin’.
Governor DeWine is first up in our Quartet of Confidence today, expressing full trust in his newly-appointed DEW Director. “Steve Dackin is someone who started as a teacher, someone who’s been a principal… He’s been a superintendent. He’s seen it also from a community college point of view, which I think is also very important.” The biggest plus for the Gov: “For the first time, when I sit down with my cabinet, an official member of the cabinet will be the superintendent of public instruction.” (Cleveland.com, 11/13/23)
The superintendent of Salem Christian Academy in Clayton, Ohio, expresses rock solid belief in this op-ed that families have the right to choose how their tax dollars are spent in educating their children, that private schools are demonstrably better options than traditional districts for many of those families, and that school choice competition improves outcomes for traditional districts. All in less than 600 words. (Dayton Daily News, 11/15/23)
Ohio policymakers have undertaken significant initiatives in recent years aimed at improving the academic outcomes of all our students. These include adopting rigorous grade-level standards in reading and math, and crafting a school report card that shines light on the progress students make toward these goals. State leaders have also steadily boosted funding for public education, and they recently passed an initiative to overhaul reading instruction in Ohio schools.
These are all important and praiseworthy efforts. Yet one under-discussed issue remains: the much-needed reform of school district governance. The report below offers recommendations aimed at enhancing local, citizen-led accountability for district performance, while also loosening bureaucratic constraints on district leaders so that they can focus on what matters most: Lifting student achievement.
Introduction
Despite astronomical graduation rates, too few Ohio students exit high school prepared for what comes next. Just 24 percent of the state’s graduating class of 2021 met college-ready standards on the ACT or SAT, and only 8 percent earned industry-recognized credentials. Data from the National Assessment of Educational Progress (NAEP)—the “Nation’s Report Card”—consistently show that fewer than half of students are reaching proficiency in math and reading. On the most recent round (2022), just 33 percent of Ohio eighth graders reached that benchmark in reading and only 29 percent were proficient in math.
Mediocre outcomes such as these have sparked great debate about how to put more students on surer pathways to success. Some public school advocates continuously assert that more money is the answer. But funding has risen while achievement has not, and that’s the case even more in the wake of the Covid pandemic. In 2021, the average Ohio school district spent more than $14,000 per pupil—up from the roughly $11,000 per-pupil average in 2000.[1] Funding will continue to escalate in FYs 2024 and 2025, as state lawmakers recently provided a hefty boost in district funding via the newish Cupp-Patterson funding formula.
Yet adding money is likely to disappoint in the absence of other serious K–12 education reforms. On that count, Ohio has taken some commendable steps forward by developing rigorous state learning standards and a transparent report card that sheds light on student progress. Lawmakers have also expanded educational choice, which empowers parents to engage more directly in their child’s education and—through heightened competition—pushes all schools to improve. Just this year, state lawmakers enacted historic policy improvements that bolster the state’s public charter and expand access to private schools; these initiatives will further advance choice and competition.
Such moves are intended in part to incentivize stronger performance in traditional public school systems. Yet efforts to directly reform the way districts are governed and managed are often left undiscussed, despite their potential to boost student outcomes, perhaps even more dramatically than the incentives noted above.
Though politically challenging, state legislators interested in such achievement boosts should consider efforts to modernize today’s school district. Teachers working in districts are too often treated like widgets—not the professionals they are—as their performance, talent, and hard work have no bearing on their pay. Job protections shield poor-performing employees from discipline or removal, inhibiting student learning and damaging organizational morale. Sprawling union contracts tie the hands of school leaders, making it nearly impossible to effectively manage their staff to meet pupil needs. To top it off, archaic election laws weaken community-driven oversight and accountability for district leadership and instead allow special interest groups—most notably local unions—to exert outsized influence. Facing little electoral pressure from parents and the general public, school boards tend to avoid pressing for much-needed changes that put students first.
Through bold reforms, state lawmakers can pave the way for district schools whose eyes are on the prize: higher student achievement. The recommendations in this report seek to address state policies that erode local accountability for performance while impeding effective management of district schools. The recommendations are based on the following principles:
Accountability: Leaders and educators at all levels—from school board to classroom—should be held accountable for increasing student learning.
Autonomy: District leaders should have the authority to develop and manage teams of talented, dedicated educators.
Excellence: Schools should value, incent, and reward strong employee performance.
In sum, state lawmakers should strive for a high-accountability, high-autonomy district sector. This approach, which some have described as “tight-loose”—tight on ends (student outcomes) but loose as to means—would be more consistent with the thinking of prominenteducationscholars and adhere to international research suggesting that nations with highly accountable yet autonomous school systems deliver better student outcomes. It would also follow studies fromvariouslocales in the U.S. that find public charter schools, which operate under this basic concept, outperform their district counterparts.
It’s true, of course, that structural changes alone are no cure-all for the inadequate performance of much of public education. Curricular and instructional reforms—some of which are already underway with the current moves to follow the science of reading—are also critical. So, too, are effective student supports, smart budgeting, and (as noted earlier) rigorous state accountability systems and quality educational choices. But a policy framework that promotes strong district governance and dynamic management remains an essential piece to the overall puzzle.
With student achievement floundering and made worse by the pandemic, there’s no time to lose. Transforming the traditional school district will be yet another step toward the education outcomes—and brighter future—that Ohio’s children need.
Holding school boards accountable
School boards play a central role in district governance. They are responsible for hiring the superintendent and treasurer, then overseeing and evaluating their performance. Boards adopt textbooks and curricula, craft and approve budgets, and decide whether to put local tax requests on the ballot. Together with administrators, they negotiate employment contracts with the unions. They set priorities and policy for the district, monitor progress, and respond to concerns from parents, employees, and community members. Despite these weighty responsibilities, board members receive little to no compensation, and some put in long hours reviewing materials, responding to constituents, attending meetings, and appearing at school events.
Although school board members are a step removed from teaching and learning, they set the tone for the district and influence what happens inside classrooms. Some do an exemplary job—exercising the type of leadership that moves the needle on pupil achievement—but others get mired in petty politics, micromanagement, and excuse making, setting the stage for an organizational culture rife with low expectations (a 2014 Fordham Institute study showed a link between more academically focused school boards and higher pupil achievement).
Given boards’ influence and authority over their districts, state lawmakers should help communities hold members accountable for improving student outcomes. Yet weaknesses in current law undermine citizen-driven accountability and shield board members from scrutiny. To strengthen accountability, we suggest the following:
Schedule school board elections on cycle. The traditional mechanism for holding school boards accountable is through the ballot box. If local voters are frustrated with district performance, they can and should select new members. The Ohio Constitution calls for “off-cycle” school board elections—i.e., taking place in odd-numbered years when national and statewide elections are off the ballot. Such timing dramatically reduces voter turnout: Just a quarter to a third of voters head to polls in off-cycle elections, whereas half to three-quarters vote in on-cycle elections. Table 1 shows the sharp differences in voter turnout across Ohio’s three largest counties. In Cuyahoga County, for instance, just 27 percent of registered voters participated in the fall 2021 off-cycle elections when school board races were contested. Meanwhile, 71 percent of the county’s voters went to the polls in fall 2020—a presidential election—and 47 percent voted in fall 2022 (midterms).
Table 1: Voter turnout rates for off-cycle (2019 and 2021) and on-cycle (2020 and 2022) electionsSource: Cuyahoga, Franklin, and Hamilton County boards of elections
Low-turnout contests tend to favor certain segments of the electorate rather than the general public. Analysts have found that off-cycle school board electionsgive undue sway to district employees who are highly motivated to elect candidates sympathetic to their concerns (e.g., higher pay or easier work conditions). Though it would require significant legislative support and, ultimately, voter approval via statewide referendum, moving board elections on cycle would change this dynamic by significantly increasing public participation and encouraging boards to pay attention to the concerns of the broader community. Accountability for board members’ decisions would also increase, as they could no longer count on sleepy elections to stay in office so long as they satisfy special interest groups whose goals are not always aligned with the best interests of students.
Include candidates’ party identification on the ballot. School board races in Ohio are technically nonpartisan under state law, and candidates’ party identification is omitted from the ballot. Without a “cue” about where candidates likely stand, the only thing a voter may know about a candidate is a name or how many yard signs he or she has—hardly sufficient information to make an informed choice. Suppressing party identification also weakens accountability for incumbents, as voters cannot identify which party holds the board majority. If candidates’ party affiliations were revealed, dissatisfied citizens could vote in a manner that seeks to shift control of the board. As occurs in Louisiana and Pennsylvania, Ohio should require party identification to appear on the ballot for school board elections, so that time-constrained citizens—not just political “insiders,” often those with union ties—have some baseline information about the candidates and their likely leanings. A primary election process, however, should not be used to nominate board candidates, thus allowing multiple candidates from each party and independents to run. Of course, state and community leaders should continue to encourage Ohioans to learn more about board candidates by attending town halls, reviewing campaign literature, and directly engaging with those running for office.
Increase modestly the maximum allowable compensation of board members. Ohio’s local school board members are currently paid very little, if anything. Statute sets a maximum compensation of just $5,000 per year. While most board members likely see the position as community service, poor compensation could easily discourage talented community members from seeking the position. It might even cost them more to serve on the board—counting time, travel, childcare, and more—than they receive in compensation. Predictably, many board races go uncontested each cycle, with incumbents unopposed or only weakly challenged. In the fall 2021 elections, four Franklin County districts had uncontested board races (three candidates vying for three seats), while three more had weakly contested races (four candidates for three seats). Increasing the maximum pay would be more consistent with the responsibilities and commitments of a board member. It could also remove a financial barrier to working-class individuals and busy parents who seek to pursue the office. One potential way of structuring board compensation is to create multiple tiers based on district enrollment and the likely commitments of the position—e.g., max compensation of $10,000 for districts with more than 10,000 students; $8,000 for districts between 5,000 and 10,000; and $6,000 for districts below 5,000 students. Florida, for instance, structures its school board compensation policy in this way, so that members of larger districts receive higher compensation than their counterparts overseeing smaller ones.
Include a statement about improving student achievement in the oath of office. Under state law, board members must affirm an oath of office that includes upholding the U.S. and Ohio constitutions as well as faithfully performing the “duties of his office.” Those are essential provisions. But, as suggested in this scholarly article, the oath should also include a statement about improving student outcomes. Pledging school board members to such a goal would make clear that one of their highest obligations is to benefit students. It could also make increasing achievement more central to board debates and reelection campaigns.
Creating fairer local tax referenda
On average, Ohio districts generate 43 percent of their total funding from local taxes, with the rest coming from state and federal aid (41 and 9 percent, respectively) and nontax revenues (7 percent). Local dollars are raised primarily via property taxes, which a third of Ohio districts supplement with local income taxes. Save for minor exceptions,[2] local voters must approve these taxes through ballot measures initiated by district boards. Ohio law requires all districts to levy a minimum twenty mill (or 2 percent) property tax rate in order to qualify for state aid, and voters in virtually all districts have approved tax rates above that floor. In 2022, districts’ effective property tax rates ranged from 2.0 to 8.3 percent, with a statewide average of 3.5 percent.
The local levy system generates billions of dollars for districts and remains a key aspect of school funding. Yet the referenda requirement also serves as an accountability check, as residents can express their support (or not) of the district through these ballot measures.
In a fair election, neither proponents nor opponents of a district tax request would receive a built-in advantage from state policy. The tax question on the ballot, for example, should be presented in neutral and understandable language. To their credit, lawmakers recently made changes to ballot language that now require a clearer presentation of the proposed cost to homeowners. Yet Ohio law still provides subtle edges to those who want higher taxes. This tilted playing field weakens accountability, as districts can rely on those advantages that make it easier to pass a levy. To create a more even playing field, we recommend the following:
Tighten requirements for an emergency levy.Ohio law allows districts to seek an “emergency levy,” one of several types of levies that districts may put on the ballot. Surprisingly, districts need not be in state-defined fiscal distress to propose this type of levy. In fact, a district need only claim that it’s seeking the dollars “to avoid an operating deficit,” regardless of the size of the projected shortfall or whether the district could balance its budget by adjusting expenditures. Roughly one-third of Ohio districts—212 of them in 2022—imposed an emergency levy, even though just two are in fiscal distress. One district not in fiscal distress is Cincinnati Public Schools, yet it passed an emergency levy in fall 2022 while at the same time spending $18,075 per pupil that year (23 percent above the state average). Districts shouldn’t be able to frighten or mislead voters into a yes vote through “emergency” terminology. Unless they are in an actual state-determined fiscal emergency, districts should be required to propose more neutrally framed funding requests.
Restrict districts to one levy request every twelve months. In tax campaigns, districts sometimes try to coax voters into approval by threatening to put the levy—should it fail—on the ballot again. Without restrictions on how often districts can go to the ballot, some follow through on their “promise” and ask voters to approve tax requests multiple times in short duration. Ross Local School District, for instance, recently put a levy on the ballot three times within one year. Preble-Shawnee Local Schools put six tax requests on the ballot within a six-year window—twice each in 2017 and 2020—before it finally passed a levy in 2021 (“sixth time’s the charm,” said the local superintendent). The unfettered ability to return to the ballot is unfair to opponents of tax referenda, and the repeated attempts can wear down voters into grudgingly approving it. State lawmakers should make sure that “no” votes—a measure of citizen-led accountability—are better honored by limiting districts to one tax request every twelve months.
Require tax rates and levy histories to be posted on a district’s report card and its own website. Ballot language discloses the amount of the proposed levy but not the total tax burden that voters face if it passes. For instance, a district may propose a property tax hike of five mills (0.5 percent), but voters have no information—at least not via the ballot—about their existing tax rate (which might already be twenty-five mills or more). While the ballot itself could include preexisting tax rates, doing so would increase the length and complexity of the question. Instead, legislators should require districts’ local tax rates (along with comparisons to state averages) and levy histories to appear in the financial section of their state report card so that voters have easier access to important background information before they head to the polls. Districts should also be required to disclose these data on their own websites.
Advancing excellence in the office of superintendent
Each school board must appoint a superintendent who, alongside the board, develops district priorities and strategic plans as well actually managing the district. The superintendent is typically the highest-paid employee of the district—often earning well into six figures—and he or she exercises leadership in a wide variety of areas. This often includes recommending curricula and instructional programs, negotiating employment contracts with the unions, forging partnerships with community organizations, and fielding concerns from educators, parents, and citizens.
Much like board members, superintendents are a step removed from the classroom, but their work strongly influences both the culture and practices of the district. Nationally, several big-city districts have benefitted from strong superintendents, some of whom came from nontraditional backgrounds such as law, politics, the military, or nonprofit management. Such leaders include Joel Klein in New York City, Arne Duncan in Chicago, Mark Roosevelt in Pittsburgh, and Mike Miles, the recently appointed superintendent in Houston.
Although advancing excellence in the superintendency is an important goal for Ohio, today’s state policies foster weakness in the central office. School boards, for instance, are not required to conduct rigorous annual evaluations of superintendents, nor can they dismiss a superintendent without jumping through a multitude of regulatory and legal hoops. There are also significant entry barriers to the job; as a result, superintendents typically come from educator ranks—a background that may or may not provide the skills and experience needed to run large, complex organizations. As former Cleveland superintendent Eric Gordon said in a recent interview, “Honestly, a superintendent’s license did not prepare me for this job. . . . So I actually had to park a lot of my academic officer skills and focus on other things, like the finances, like talent, like the infrastructure, the organizational structure, public trust, the politics, all of these other things that they didn’t teach me in superintendent school.”
To create the potential for boards to engage real “chief executive officers” for their districts, we propose the following:[3]
Remove unnecessary licensure restrictions. Current state law requires district boards to hire individuals with a superintendent’s certificate, the requirements for which are extremely restrictive. For example, as a prerequisite for a traditional superintendent license, an individual must hold a master’s degree and have previous experience being a teacher as well as a principal or central administrator. These criteria all but eliminate the possibility of hiring superintendents from other fields and backgrounds. Ohio does allow school boards to hire nontraditional superintendents under a temporary, two-year alternative license that can be renewed once. But that route includes its own onerous requirements, such as completing additional coursework to move to a permanent license, along with time-consuming professional development while serving under the alternative license. Those hurdles likely discourage external leaders from considering the job—and boards from seeking them. State lawmakers should revamp superintendent licensing to allow districts to more easily hire nontraditional candidates. Individuals with a master’s degree or higher and with at least ten years of professional experience should be allowed to apply for a standard superintendent license. This would permit school boards to hire professionals with MBAs, JDs, and PhDs as their superintendent, without any further requirements. University governing boards sometimes hire “outsiders” as president. Ohio State University, for example, just hired a decorated military veteran, and the University of Florida recently hired a former U.S. senator. Why shouldn’t school boards be able to pursue a leader with those types of backgrounds?
Allow district boards to more easily dismiss a superintendent. Many top government and private-sector leaders serve at the pleasure of their appointing authority and can be dismissed without formal justification. But this is not the case with district superintendents in the Buckeye State. While a school board may terminate a superintendent’s contract before it expires, it must do so for “good and just cause.” Though this may seem sensible on its face, the provision allows for long and expensive legal proceedings that could pit the board against a dismissed superintendent who believes his or her firing is not “good and just.” These job protections for superintendents weaken boards’ authority and discourage removal, even if they have serious concerns about a superintendent’s leadership. State lawmakers should make clear that boards can remove superintendents without undue legal burdens, in essence treating superintendents as an at-will employee.[4]
Require annual superintendent evaluations that include consideration of student achievement—and make evaluations public. Current state law requires district boards to have some type of evaluation process for the superintendent, but how and when that’s done are left entirely to the discretion of the board. A recent news story from Southwest Ohio indicates that some boards do perfunctory reviews of superintendents—one district had a one-paragraph evaluation—and a few didn’t even bother to complete a written evaluation. That’s just unacceptable. State lawmakers should ensure that boards adopt an evaluation that includes an academic element based on student outcomes (though specific targets should be board determined). Legislators should also clarify that boards must conduct an annual evaluation. Finally, as a matter of transparency, boards should be required to make public the evaluation rubric as well as the results of their evaluations. This would hold boards more accountable for implementing a rigorous review while also putting some healthy pressure on the superintendent to perform.
Letting district leaders lead
In Jim Collins’s classic management book Good to Great, he argues that the first and most critical job of organizational leaders is to get the right people on the bus—and the wrong people off. He explains, “Your best ‘strategy’ is to have a busload of people who can adapt to and perform brilliantly no matter what comes next.” That’s wise advice, not only in business and industry but also for education leaders.
Unfortunately, a complicated web of Ohio state policy constrains district leaders’ ability to get the right people aboard. Rigid, seniority-driven salary schedules handcuff leaders who wish to pay competitive salaries in an effort to attract early- and mid-career professionals, to reward their finest teachers, and/or to persuade them to take the toughest assignments in the most challenging schools. Unwieldy administrative hurdles make dismissing poor performers arduous and expensive. Union contracts, which can cover almost any issue under the sun, make it hard to maintain workplace discipline and build a professional, results-driven organizational culture.
Rather than tying their hands, state lawmakers should empower district leaders to effectively manage their teams. Ohio law, after all, declares that school boards “shall have the management and control of all of the public schools . . . that it operates in its respective district.” Less prescriptive staffing regulations would not only follow this legal precept but also align more closely with the management practices found in other realms of society. It would also likely find support among Ohio’s district superintendents, as strong majorities expressed support for more flexible HR policies in a 2011 Fordham survey. Though not entirely successful, former Governor Kasich pushed hard for serious reforms to school management, and state leaders continue to voice interest in—and have taken some steps toward—giving public schools greater flexibility in hiring and staffing.
Most importantly, better school management is likely to improve student outcomes. Studies from Wisconsin and Florida find that achievement rose when district leaders were given more authority over teacher compensation (WI) and dismissal (FL). Public charter schools are freed from some of the burdensome staffing regulations that districts face, and there is strong evidence from across the nation that charters outperform their less-autonomous district counterparts. To free district leaders to effectively manage the workforce—and to promote a culture of high performance—lawmakers should do the following.
Repeal districts’ ability to waive management rights and narrow the scope of collective bargaining. Ohio law delineates specific “inherent” management rights of district leadership, which include the ability to “direct, supervise, evaluate, or hire employees” and “effectively manage the workforce.” Yet that same section of law allows districts to give away these rights through the collective-bargaining process, and it also permits negotiations over the “terms and conditions of employment” (in addition to wages and hours). Taken together, these provisions open the door to haggling with unions over myriad workplace issues, including class size, employee transfer and attendance policies, evaluation procedures, disciplinary processes, and more. Dayton Public Schools’ contract, for instance, outlines extensive formal disciplinary procedures that school leaders must follow if they seek to address employee misconduct—red tape that could easily discourage them from taking corrective action.[5] District leadership should ultimately be responsible for establishing workplace rules and job expectations. To ensure this happens, state lawmakers should remove the provision allowing district boards to agree to policies that erode managerial authority. They should also narrow the scope of collective bargaining to wages and hours, as do states such as Indiana and Montana.[6] Such changes would make clear that basic management responsibilities are nonnegotiable.
Eliminate mandatory step-and-lane teacher-salary schedules. Ohio law requires districts to adopt salary schedules based on teachers’ seniority and college credits earned. Hence, districts must pay their senior teachers more even if they are not as effective or are teaching in less demanding subjects or schools. This policy also mandates substantial salary bumps when teachers receive a master’s degree, despite the scant evidencethat a master’s degree enhances effectiveness. This pay system thus ignores an individual teacher’s contributions to student learning, specialized skills, and willingness to work in more challenging environments. It also severely limits district leaders’ ability to recruit and retain talented young people or mid-career professionals, as they are forced to begin their careers at the bottom of the pay scale and are locked into meager annual raises thereafter. To address these issues, state law should be silent on teacher salary[7] and permit district leaders to pay teachers based on merit and other important factors unrelated to seniority or credits earned. Districts could continue a step-and-lane pay policy, but that would be by choice rather than compulsion.
Remove coursework requirements for teacher tenure and instead make strong performance evaluations a condition for eligibility.Statute outlines several conditions that teachers must meet to be eligible for tenure—i.e., employed under a “continuing contract” that provides significant job protections during their entire career.[8] In addition to holding an educator license for at least seven years, most teachers[9] must complete a whopping thirty semester hours of additional coursework to be eligible for tenure. Like the credits-earned dimension of the salary schedule, this is another perverse incentive for teachers to sit through more coursework and pursue expensive credentials that don’t clearly benefit students. Instead of tying tenure to coursework, state lawmakers should make satisfactory evaluations a condition of tenure eligibility. As recommended by the National Council on Teacher Quality and implemented by states such as Michigan and Tennessee (and to good effect in the Volunteer State, per one recent study), Ohio should move to performance-based criteria for tenure. In addition to finishing a seven-year probationary period—a lengthy term that Ohio lawmakers, to their credit, put in place in 2009—nontenured teachers should need to receive a rating of “skilled” or “accomplished” (the two highest ratings) on their two most recent years of teaching to be eligible for tenure. This approach would ensure that teachers demonstrate instructional effectiveness before receiving job protections.
Make dismissal of low-performing principals and teachers less onerous. Much like superintendents, school principals and tenured teachers are entitled to extensive hearings and appeals processes that may require their employers to demonstrate “good and just cause” for a dismissal. These administrative hoops—faced neither by most private employers nor by most charter and private schools—discourage district leaders from dismissing ineffective employees, as it’s not absolutely clear that poor performance constitutes “good and just cause.” Following the lead of states such as Florida and New York, Ohio legislators should clearly articulate that ineffective leadership and instruction is grounds for termination, thus removing the possibility that a district’s decision is overturned on appeal. State lawmakers could, in fact, simply extend to all districts a statute that currently provides the Cleveland school board with “just cause” to terminate a principal or tenured teacher who receives poor evaluations.[10] Districts could still be required to demonstrate cause for other reasons, such as alleged misconduct. But making sure that poor performance is clear grounds for dismissal would provide stronger backing to district leaders seeking to remove ineffective principals and teachers, while also putting teeth into the state’s principal and teacher evaluation systems.
Eliminate last-in, first-out provisions based on tenure status. Tenured teachers also have job protections in cases where a district pursues reductions in force because of enrollment declines or financial reasons. Under a last-in, first-out type of policy, districts are forced to lay off nontenured teachers before those with tenure.[11] This continues a “quality-blind” approach to layoffs, as performance is not considered in the first (and perhaps only) cut. Only tenure status matters. Instead of basing layoffs on tenure, lawmakers should require districts to first lay off “ineffective” teachers within a teaching field—those who receive the lowest rating in their annual evaluations—and then move to “developing” teachers, the second-lowest mark. This would allow district leaders to retain effective teachers, regardless of their tenure status.
Conclusion
Traditional public school systems continue to educate a large majority of Ohio students, yet too many students still leave high school ill prepared for their next step in life. Education reformers have long insisted that one factor behind these mediocre outcomes is a district governance model that tends to put adult school employees’ interests above parent and student needs. They are spot on, and the urgent need for change is even felt by those leading public school systems. In a 2011 survey of Ohio superintendents, strong majorities expressed support for measures such as repealing step-and-lane teacher-pay systems and reigning in collective bargaining. In a scorching resignation letter from 2018, a former Columbus City Schools board member stated quite plainly that the district governance structure is “completely inadequate.”
Regrettably, not much has changed since then. Collective bargaining over all sorts of workplace issues is still a fact of life in Ohio school districts. Teacher pay is still determined via lockstep salary schedules. The pathway to the superintendents’ office still runs through a conventional educator career “ladder.” Local, citizen-led accountability for district performance remains weak due to misguided election laws.
It’s time for a change, and Ohio lawmakers can promote a stronger, more student-centered public school system by making overdue changes in statute. Give district leaders the autonomy and freedom needed to lead, but hold them more accountable for results. Doing so would benefit not only district employees, the vast majority of whom want to excel and accomplish great things, but also the 1.5 million Ohio students who attend district schools.
Appendix
Table A1: Ohio statutes referred to in the report
Acknowledgments
I wish to thank my Fordham Institute colleagues Michael J. Petrilli, Chester E. Finn, Jr., Chad L. Aldis, and Jessica Poiner for their thoughtful feedback during the drafting process. Jeff Murray assisted with report production and dissemination. Special thanks to Pamela Tatz who copy edited the manuscript and Andy Kittles who created the design.
- Aaron Churchill
Endnotes
[1]The 2000 spending figure is adjusted for inflation to 2021 dollars using the consumer price index.
[2]For instance, school districts receive a certain portion of municipalities’ inside millage—i.e., unvoted property taxes that local municipalities may levy at rates below ten mills.
[3]School-level principals are, of course, another key leadership position. The way they are hired, licensed, and held accountable is a substantial topic that—apart for some discussion on page 9—is saved for another day.
[4]At-will employees in any sector have certain job protections that allow them to appeal terminations based on issues such as discrimination or retaliation. Those protections should certainly apply if superintendents become at-will employees of the board.
[5]Page 85–86 of the Dayton Public Schools’ teachers’ contract outline disciplinary procedures.
[6]Four states—North Carolina, South Carolina, Texas, and Virginia—do not allow collective bargaining for teachers. Wisconsin allows bargaining only on wages.
[7]In addition to the salary schedule, state law also restricts districts’ ability to reduce the annual pay of an individual teacher; pay reductions must be done as part of a system-wide reduction. That language should also be repealed, potentially allowing district leaders to use reduced pay as a disciplinary measure or as a way to retain a teacher who would otherwise be laid off due to budget cuts.
[8]Nontenured teachers are employed under “limited contracts,” with maximum terms of five years and which districts may nonrenew. In cases of nonrenewal, employees would be let go and have less recourse for reinstatement through appeals. In contrast, a “continuing contract” does not expire unless a teacher retires, resigns, or is terminated.
[9]The coursework requirement applies to teachers initially licensed (after 2010) without having earned a master’s degree. This likely applies to the vast majority of teachers who enter with a bachelor’s degree as their highest credential.
[10]In Cleveland, the district has just cause for dismissing a tenured teacher who receives an “ineffective” rating—the lowest mark—for two consecutive years, while principals may be dismissed for “failure for the principal’s building to meet academic performance standards established by the CEO.”
[11]Teachers’ tenure statuses are first compared within their teaching field to make a first round of cuts. Seniority cannot be used as a factor unless a board is comparing teachers of “comparable” evaluations; e.g., two nontenured math teachers with the same (or similar) rating, or two tenured math teachers with the same (or similar) rating, after nontenured personnel have been exhausted.
Editor's note: This was first published on the author's Substack, The Education Daly.
You would be hard pressed to find a better encapsulation of grade inflation than the graph below. It comes from the Harvard student newspaper’s survey of 2022 graduates. The purple band on the far left, which includes just under 9 percent of Harvard’s most struggling students, is for those with grade point averages of 3.5 or below. Everyone else—91 percent of the senior class—reported earning a 3.6 or higher.
It looks bonkers. How can all of Harvard’s students possibly be earning such high grades? Do we even need six different categories in this graph? Aren’t we just distinguishing the kids with all A’s from their peers who scraped along with almost all A’s?
Many people find this extremely frustrating. To them, privileged Ivy League kids are being handed grades they could not possibly be earning, which will in turn guarantee them admission to the best graduate schools and job offers from the most prestigious companies. Harvard sucks! Go ahead, say it and get it out of your system so we can move on.
This narrative about college grade inflation goes back decades—at least to the 1960s. The trends are impossible to deny. Grades rose much faster than achievement—particularly at private institutions. Professors became more willing to hand out A’s and less likely to give C’s or anything lower. It has been widely reported in the press, researched in depth, and addressed by blue-ribbon committees. I won’t bore you with the details. Bottom line: None of it made any difference. Grade inflation marched onward and upward.
That problem is not my focus in this post.
For me, grade inflation isn’t important as it relates to Harvard students but as it relates to low-income students in our K–12 schools. When they receive inflated grades, it’s not a windfall of unearned currency that opens doors to future success. Instead, grade inflation often robs them and their families of critical opportunities by giving them falsely reassuring messages. It’s not a victimless crime. In fact, it’s one of the most important—and under-discussed—issues in education today.
In part 1, I lay out the basic shape of the problem. In part 2, I will consider possible fixes.
What is grade inflation?
We need a definition. Grade inflation is the tendency to award increasingly positive marks to students without any corollary improvement in the quality of work or student performance.
It often makes grade descriptors sound ridiculous. For instance, some school systems define a B grade as “above average.” But if 90 percent of students are earning a B or better, that’s weird because 90 percent of students can’t be above average. Garrison Keillor used to joke about this. Grade inflation.
Is grade inflation actually a problem?
Not everyone thinks so.
For almost as long as curmudgeons have been lamenting grade inflation and ordering pampered younger generations to get off their academic lawn, there have also been rebuttals from those who feel it’s not a real crisis. They have some good points that are worth considering. Here are a few:
Higher grades increase kids’ confidence and encourage them to persist. There’s a really interesting paper by two professors, Zachary Bleemer and Aashish Mehta, who considered what happens when a college only allows students to major in a rigorous subject—economics—if they earn a 2.8 GPA in their introductory economics courses. Quite a reasonable policy, right? The students who fall short of the 2.8 generally choose a less rigorous major...and end up earning substantially less income in their careers. But the researchers found that those students who were prevented from sticking with economics generally would have succeeded if they were allowed to continue—and they would have gotten higher-paying jobs. A bit of grade inflation to get them over the 2.8 GPA bar would have been a productive thing, they argue.
How persuasive is this argument? It’s provocative. It’s also a stronger case against having a minimum GPA requirement for economics than a case forgrade inflation, but if we’re trying to diversify the pool of kids who earn economics degrees, we shouldn’t be pushing out kids who could succeed. Fair enough.
Even when inflated, grades make it possible to distinguish the strongest students from peers. According to this perspective, grading scales are arbitrary. There is no shared definition of an A. No matter which system is used, the top performers get the highest marks and the lowest performers get the worst. Schools can rank students or bestow honors and awards on those who have earned them. Going back to our Harvard example above, the 4.0s stand apart from the 3.7s. As long as there remains a discernible hierarchy, grade inflation isn’t a real issue.
How persuasive is this argument? If we are using grades to compare students across an entire grade or school, this argument makes sense. But it sees the world through the eyes of an administrator. What about parents? They have no clue what grades the other students are earning. They rely on grades to tell them whether their kids are thriving academically. If they earn A’s, they assume those A’s mean the same thing they did back when the parent was in school. In these cases, inflation leads to distorted messages. We will come back to this issue. It’s a key part of the whole picture.
There’s new evidence about grade inflation for us to consider
More lenient grading seems to contribute to absenteeism. Wait, really? Yes. Researchers recently found that when North Carolina changed its policies for high school students so it became easier to earn good grades, struggling students began missing more school—while there was no similar increase in absenteeism for high performers. Grade inflation widened performance and attendance gaps. Given that we are trying to reverse a national epidemic of chronic absenteeism, we should be particularly worried by these findings. They show that not all students have the same response to lenient grading. Thriving students gobble up the better grades like Pac Man pellets when it is easier to get them. Other kids are content to earn the same mediocre grades while investing less effort.
Student achievement fell significantly. The average student is five months further behind in math and English language arts than was the case before the pandemic.
Chronic absenteeism went through the roof. About one in five students is missing at least 10 percent of school days—or eighteen days per year. This figure has more than doubled in both districts compared to 2018.
Yet most students are still earning the same grades—or even better grades.
That’s a really concerning trifecta: lower performance, worse attendance...and grades that didn’t change. As of 2021–22, about 40 percent of the students in these two districts who were chronically absent and not performing on grade level according to state tests nonetheless earned grades of B or better.
No wonder American parents are convinced that their kids have already recovered from any pandemic-era setbacks.
If report cards have contained nothing but happy news, why would parents have any reason to believe there had ever been any setbacks? Why would they register their kids for supplemental tutoring? Or voluntarily send them to summer school? They wouldn’t—and this is a big reason why our pandemic recovery is not going very well. In many states, students—particularly those from less privileged homes—are still lagging behind their 2019 performance levels.
These patterns of pandemic grade inflation were widespread.Also released today, Dan Goldhaber and Maia Goodman Young reviewed nearly a decade of middle and high school grading data for math, science, and English courses in Washington State. In their paper, they find that after state officials directed district leaders beginning in March 2020 to ease grading standards, GPAs in all three subjects rose sharply—and did not return to pre-pandemic levels even during the 2021–22 school year.
Importantly, Goldhaber and Goodman Young show that the connection between grades and student performance on state tests—particularly in math—weakened. Inflated grades became a fuzzier achievement message for parents. In this case, we’re talking about an entire state rather than the two districts we featured in False Signals.
Alyssa Rosenberg has a very nice column summarizing the studies and implications in the Washington Post.
But wait… there’s more! Tom Swiderski and Sarah Crittenden Fuller also recently published a paper on North Carolina that echoes, almost word for word, the findings in Washington State. They suggest that the “growing gap between student GPAs and achievement could be contributing to parents’ confusion about the extent of their children’s needs for pandemic recovery supports and an under-utilization of recovery programs.”
Truly an avalanche of research.
Although these policies reflect genuine empathy for students and families during a difficult time, we are seeing a rapid accumulation of reasons to believe that grade inflation is harming the very students it was intended to help.
—
In part 2, we will look at options for addressing all this grade inflation with an emphasis on practicality. Principals firing off memos ordering their teachers to grade harder is not going to work here.
Fortunately, some smart folks from a wide array of perspectives have thought about this problem. We will see what we can learn from them. I’ll add my own recommendations.
Editor’s note: This essay is an entry in Fordham’s 2023 Wonkathon, which asked contributors to answer this question: “How can we harness the power but mitigate the risks of artificial intelligence in our schools?” Learn more.
Many of my educator colleagues have recently expressed concern about the rise of AI in our classrooms. I haven’t agreed. In fact, AI programs like ChatGPT do not worry me. I’m increasingly convinced that some futuristic version of education will not replace our classrooms, nor our teachers. And the precise reason I am not worried is because I feel as though AI has already been tried in our classrooms.
Hasn’t it?
I mean what do we call a decades-long educational system that eliminates recess, sports, art, history, and science for many marginalized community schools in the name of higher reading and math scores? Or the National Reading Panel‘s finding in 2001 that reading for pleasure does not impact “achievement”? What would you call the literal scripting of standards-based instruction around key standards and test preparation?
More recently, how would we describe a virtual classroom where one non-credentialed teacher monitors nearly a hundred kids plugged into an online curriculum?
If these are not intelligences of an artificial nature, I do not know what is.
But why did these systems of AI fail? Why are they failing now when put into practice? That answer is simple and needs no coding nor algorithm. It’s because kids are not computers, and what we must teach them to be healthy, happy, and well-educated adults has never been less artificial.
AI cannot teach compassion
During the pandemic, a wave of articles and books found publication about the need for social and emotional learning. But what the general public may not realize was that, just prior to Covid, legislators were not interested in social and emotional learning. I know, as I was a policy fellow trying unsuccessfully to find champions for a bill in California to fund SEL. But then there was Covid, and kids in front of computers expanded the SEL conversation. Suddenly, my colleagues and I found ourselves in state senators’ offices instead of in the hallway talking to staff.
Why? That’s because parents and educators alike now see what limits technology like AI actually has. Parents want their children to grow in their social and emotional intelligence, not just academics. Computers simply cannot teach compassion, and despite well-advertised efforts, they never will.
AI cannot teach work ethic
During the pandemic, realities on the computer made work completion optional in district after district. Soon, educators and educational writers were noting a brand-new term, too—learning loss. That’s because it wasn’t just a loss of access to technology during Covid. If that had been the case, only students not on their computers during class would have experienced learning loss. The simple fact is that AI cannot, nor can any technology by itself, replace a human teacher instilling the values of a work ethic in their students.
AI cannot teach stewardship for the environment
Climate change and STEAM science is at the top of many priorities, nationwide. It’s not just for employment sakes either.
AI can relegate tasks to students about environmental issues. AI can teach and even assess content about environmental issues. But AI cannot teach the type of stewardship for the environment that real humans do in various organizations nationwide. That takes real humans, in the very real and natural world, giving hands-on experiences. Computer screens do not have hands.
AI cannot teach the appreciation of music, theater, dance, or any other art
As a recent colleague told me, “Art is what brings them to school.” I couldn’t agree more. My best attended classes during Covid were not mine, they were with our PE coach and our partners in STEAM who gave kids a chance to do hands-on science at home. But why can’t AI just teach these things somehow, someway? That’s because AI operates in the virtual world, and art is something that we experience in the real, tactile one. We can produce art virtually, but the experience of it requires a physical presence. And not just by ourselves. Art requires a shared experience from another real human being. Whether they are the watcher, the dance partner, the viewer, or the eater, art requires not a person and an AI; it requires people in the plural sharing their feelings about the art. AI has no feelings.
AI will not be equitable, nor will it affirm anyone’s identity or culture or empower anyone
In order for AI to do a modern teacher’s job, it would have to complete all of the impossible tasks that a real human does just by being human. But there is more required because a teacher must also consider equity, culture, and identity in their classroom.
In Race After Technology, by Ruha Benjamin, the author gives compelling evidence for why technology, including AI, can’t do our job humanely. She expertly illustrates that technology’s track record is not something anyone should be happy about, whether you are in a minority or a majority. Additionally, AI does not, nor ever will, possess the human ability to understand our differences or the ability to understand why those differences are beneficial and, often, lead to bias that benefits one group over another.
I feel very sure that artificial intelligence isn’t going to teach our kids anytime soon. Unless, that is, we haven’t learned from our earlier experiments with turning children into testing robots. That part does scare me, though. Because if we no longer prioritize very real human qualities like appreciation, equity, stewardship, perseverance,and compassion, somebody, somewhere, wanting to earn a buck is going to give it a try. And it won’t take long until we terminate what is best in us as a society.
But the good news is that to do that would be a very human decision—one which we, not AIs, are in control of.
Annie Morrison
11.14.2023
Editor’s note: This essay is an entry in Fordham’s 2023 Wonkathon, which asked contributors to answer this question: “How can we harness the power but mitigate the risks of artificial intelligence in our schools?” Learn more.
Raise your hand if you have ever sat through shoddy professional learning... Now, I bet every educator reading this would raise their hand in agreement. Here is an alarming statistic: We spend nearly $18 billion a year on professional learning in this country, yet according to Rivet Education’s research of the professional learning marketplace, only one in three teachers find the professional learning they receive to be useful.[1] That means nearly $11 billion spent on professional learning is wasted; this is equivalent to giving every teacher in the U.S. a stipend of $3,000. Let that sink in for a moment.
And we get it. Planning and executing high-quality professional learning is hard work. Whether provided internally by school or district leaders or externally by a professional learning organization—professional learning takes time and expertise to make it effective. Similarly, teachers who create their lesson plans from scratch hope that what they teach their students is aligned with grade-level standards. Education leaders who plan professional learning also have to cross their fingers in hopes that the training they develop will actually improve teacher practice and student outcomes.
What can further complicate matters is that, as a whole, the field of education lacks reliable measures to assess the quality and impact of professional learning. In October, the Research Partnership for Professional Learning (RPPL) at the Annenberg Institute released its latest white paper. It found that “professional learning practitioners agree the field lacks the tools needed to evaluate the implementation and outcomes of professional learning programs effectively.”
At Rivet Education, we know all too well the importance of evaluating the quality of professional learning. Rivet Education’s Framework for Curriculum-Based Professional Learning defines for educators the characteristics, types, and structures that construct high-quality, curriculum-based professional learning (CBPL). Our flagship tool, the Professional Learning Partner Guide (PLPG), is the only evaluation process in the country that assesses the quality of CBPL services. While the PLPG serves as the industry standard for evaluating CBPL, the evaluation process is time-consuming (upwards of 500 hours per review cycle) and currently only reviews professional learning materials, not the live delivery of the training.
But what if we can harness the power of artificial intelligence to increase the efficiency and effectiveness of measuring the quality of professional learning?
Using AI to measure professional learning quality in school districts
Imagine you are a math supervisor in a mid-size school district and your Chief Academic Officer has tasked you with creating a two-day workshop to launch Illustrative Mathematics (IM) with teachers in grades 6–8 in your district. This is the only training teachers will receive until October when your district typically hosts a professional development day, and the central office couldn’t afford to have IM come in to provide the training themselves. The problem is that you are also new to IM. Your only experience up to this point with a comprehensive math curriculum was with one that was not high quality nor aligned with your state’s standards. So, you design the training, hoping you have done enough to adequately prepare your teachers to use the new curriculum on the first day of school.
A generative AI application could be used in this scenario to review the scope and sequence of the workshop and provide the math supervisor with feedback on how their training will prepare teachers to teach IM successfully. Here is an example of an AI prompt the math supervisor could use to eliminate some of the guesswork of planning effective professional learning:
Here are the training materials for a two-day workshop designed to help teachers understand the design principles and arc of learning of Illustrative Mathematics. Suggest three ways I could improve them to help teachers new to the curriculum use it more effectively.
Using AI to measure the quality of professional learning organizations
Professional learning organizations and those who evaluate their services, like Rivet, can also use AI to evaluate the quality of their PL. Consider Rivet’s evaluation process for the Professional Learning Partner Guide (PLPG). This involves teams of expert reviewers evaluating applicants’ materials, such as presentations, handouts, guidebooks, course syllabi, coaching notes, etc. These are compared alongside Rivet’s criteria to assess if the organization provides significant evidence of robust, HQIM-aligned professional learning services. Professional learning partners that meet these criteria are profiled in the PLPG. While we are confident that our process holds a high bar for quality services, we also recognize that factors such as the quality of the facilitator in the room impact PL effectiveness.
As an organization deeply valuing quality, we asked ourselves: What if AI could review and rate the quality of in-person professional learning and reduce the human capital burden needed to observe this training in person? This would require Rivet to feed AI a lot of information to replicate the knowledge and experience of our expert review team, but it is an idea worth pursuing. Here is an example of what it could look like:
Here is the transcript from a live professional learning workshop from Pelican PL Services on internalizing unit 2 of EL Education. Can you compare the transcript to the professional learning materials submitted by Pelican PL services for this training to determine if the facilitator delivered the session as intended? Then, can you compare the transcript to the Ongoing for Teachers-specific Indicators for High-Quality Professional Learning in the Professional Learning Partner Guide’s scoring and evidence guide to determine if the workshop would receive a passing score of 7 out of 10 points, meeting Rivet’s definition of high-quality professional learning?
While both prompts provide examples of how AI can be leveraged to determine the quality of professional learning, the full extent of AI’s role in K–12 education is uncertain. With AI’s potential comes both excitement and trepidation. At Rivet Education, we think AI can play a valuable role in filling in gaps in professional learning that currently exist in the market. We have a tool at our fingertips that can eliminate the guesswork and uncertainty in planning and delivering professional learning, quickly and succinctly measure its effectiveness, and foster growth and innovation in schools nationwide. It is up to us to unlock its potential.
Education policy plays a crucial role in determining whether students reach their full academic potential. The funding systems that allocate dollars to schools, data systems that track student outcomes over time, and the policies that hold adults accountable for meeting the needs of students are all incredibly important.
But they’re also incredibly complex. For the most part, this complexity is necessary in working towards systems that are as fair, equitable, and efficient as possible. But sometimes policy can become too complex. Regulations can become burdensome instead of helpful, and changing times call for changing rules. It’s important for policymakers to take a step back every few years and reconsider what’s on the books.
In early October, state lawmakers started that process by introducing Senate Bill 168, legislation aimed at “education regulation reform.” The bill eliminates a handful of “obsolete” provisions from law and adjusts several others. It also has the support of school administrator groups, whose members believe that it would “reduce burdensome and unnecessary regulations, vest more decision-making back into local communities that know their students best, and better equip our school leaders to prepare our students for the future.”
Reducing unnecessary regulations and empowering local leaders are worthy goals. But there are a few provisions in Senate Bill 168 that lawmakers should consider tweaking before they give their final stamp of approval. Here are three areas that deserve a second look.
1. Exempting high-performing districts from certain statutory requirements
In 2016, the legislature passedan initial education deregulation bill that exempted qualifying districts from a variety of statutory requirements for a three-year period. To be eligible, districts had to earn at least 85 percent of the total possible performance index score points on their state report cards, have a four-year graduation rate of at least 93 percent, and a five-year rate of at least 95 percent.[1] The exemptions included requirements related to teacher licensure and qualifications, as well as class size. Senate Bill 168 would permit districts to renew these exemptions every three years, as long as they continue to meet the qualification benchmarks (current law does not address renewal). It would also require the Department of Education and Workforce to annually notify districts of their eligibility, and of the exemptions that exist.
At first glance, these changes seem perfectly reasonable. But a lot has changed since 2016, and it’s important to reevaluate the original policy before expanding it. For example, under the 2016 law, eligible districts have been exempt from the teacher qualification requirements of the third grade reading guarantee. More specifically, they’re exempt from provisions requiring that retained students be assigned to a highly-effective teacher or one with specialized training in teaching reading. Under recent revisions to state law, parents can now request that students be promoted to fourth grade regardless of whether they read at grade level. But even in this case, promoted students are required to receive intensive reading instruction “in the same manner” as retained students. That means they also must be assigned to highly-effective or specially trained teachers. Given the state’s efforts to prioritize early literacy instruction, as well as evidence that many teachers don’t have the training they need to effectively teach reading, it seems unwise to exempt any district, high-performing or otherwise, from this requirement.
2. Reductions in non-teaching staff
Non-teaching employees play a critical role in districts and schools. But under current law, districts making reductions to their non-teaching staff must give preference to employees under continuing contracts—also known as tenure—and then to employees on the basis of seniority. This rule ties the hands of district administrators when they need to reduce staff, as seniority can override their ability to retain high-performers with less seniority. Senate Bill 168 seeks to change this by eliminating the preference for retaining non-teaching staff on the basis of seniority unless the district is deciding between two employees with comparable evaluations.
This is a smart change, as it would give districts more leeway to retain their best non-teaching employees rather than just those who have been around the longest. It also aligns with current law regarding teachers: Districts needing to make reductions in their teaching force are prohibited from giving preference based on seniority unless they’re deciding between teachers with comparable evaluations.
But both current law and SB 168 still require districts to give preference to teaching and non-teaching staff who have tenure. And in Ohio, tenure is more or less determined by how long a staff member has been around (and, in the case of teachers, graduate coursework). It’s not tied to quality or effectiveness, meaning it’s basically another measure of seniority. If lawmakers are truly committed to empowering district leaders with regard to personnel decisions, then they should allow them to retain their most valuable teaching and non-teaching staff, regardless of seniority and tenure, when reductions in force are necessary.
3. Employing unlicensed teachers
Policymakers have recently launchedand passed a variety of initiatives and policies aimed at bolstering the teacher pipeline. SB 168 offers another. It would allow public schools to employ individuals as teachers who are not certified or licensed as long as they hold a master’s degree from an accredited institution and have successfully completed an exam in the subject area in which they will teach.
The idea behind this provision is understandable: It could expand the pool of potential teachers, while still ensuring they have some level of content knowledge. But the type of master’s degree someone has should matter, as should the age of potential students. It makes sense to allow someone with a master’s degree in chemistry to teach high school students chemistry. But someone with a master’s degree in communications shouldn’t be teaching sixth grade math, nor should someone with a master’s degree in exercise science teach third grade reading. With this in mind, lawmakers should consider tweaking the bill in two ways. First, unlicensed teachers should have to hold a master’s degree in the subject area in which they will teach in addition to passing a content exam. Second, they should be limited to teaching high school grade levels, where high levels of content expertise are most needed.
***
Lawmakers deserve kudos for tackling education deregulation, especially in areas that have been ripe for reform for nearly a decade. It’s important to ensure that district and school leaders have flexibility, and that they aren’t prevented from doing what’s best for students because of burdensome regulations. With a few key tweaks, SB 168 could be a good first step toward doing just that.
[1] Until the 2021–22 school year, districts also had to earn an A grade for performance indicators met.
Real time classroom observations by trained evaluators hold promise to accurately assess the quality of teaching and learning going on inside those four walls; an as-yet-untapped area of “education R&D”. But numerousroadblocks—lack of time and training, disagreement over rubrics, and teacher resistance among them—stand in the way of a systematized and smoothly-functioning assessment system. A group of researchers from Australia has developed a new framework that evaluates teacher and student engagement with course content that can be utilized while observing video-recorded lessons. A recent report explains their methods and describes the findings of its application.
A précis on ICAP’s four modes of engagement: “Passive” cognitive engagement is where learners are being oriented toward and receiving information from instructional materials, such as listening to a lecture or watching a video, without overtly doing anything observable related to more active learning. “Active” engagement requires focused attention while manipulating lesson materials or input—think underlining text or writing a summary of an essay. “Constructive” engagement refers to behavior that produces new ideas that go beyond the information given; students might relate new information to previous knowledge, generate inferences that are not explicitly stated in the text, or provide justifications that make the text or the problem solution more explicit. And “interactive” engagement is a group activity that meets two criteria: (a) the partners’ utterances must be primarily Constructive, and (b) the interaction must extend the generative nature of the prior contributions of the individual partners. In short, it’s a constructive-type engagement with a partner or two rather than the teacher.
Chi’s work argues three that there is a clear hierarchy among these four mode, with passive at the bottom and interactive at the top. Active engagement leads to stronger learning outcomes than passive, she says, because students engage with the material, relate it to prior knowledge, and store it in easily-retrievable ways. Constructive is better than active because students must provide explanations, raise critical questions, or complete other tasks that engage the highest levels of cognitive processing. And the interactive mode of engagement leads to the strongest learning outcomes because dialogue between learners can give rise to knowledge change processes and can create new knowledge that the partners could not have generated alone.
The Australian researchers devised an ICAP-based coding framework that distinguishes the four modes of cognitive engagement, provides reviewers clear operational criteria for defining the different modes when they observe them, and links the different modes to distinct learning outcomes. Best of all, the coding can be done using recorded lessons and transcripts, minimizing classroom disruption that can occur during live observation and maximizing the number observations a single reviewer can complete.
They watched and coded one thirty-minute video recorded lesson from twenty teachers from eight schools in the Greater Adelaide region of South Australia. All but one of the teachers taught in public schools, and all but one at the secondary level; each had more than five years of teaching experience. All had Bachelor degrees, and four also had master’s degrees. All of the teachers taught STEM-related subjects, including math, hard science (physics, biology, etc.), and soft science (psychology, health, etc.). Most schools were rated “less advantaged” via the socio-economic criteria used by the South Australian state government, including several landing at the most disadvantaged level (7 on a 1–7 scale). A few schools were on the “more advantaged” side of the spectrum, but no higher than a 4 on the scale.
The observers coded the lessons and tasks the teachers presented and assigned, as well as the verbal instructions they used in whole class directions, to determine which of the four areas of student cognitive engagement they utilized. Student actions, talk, and outputs were also examined to assess whether the actual mode of engagement that students displayed matched what was described in the teachers’ intentions for each lesson or task. Inter-rater reliability testing showed strong correspondence between different observers’ coding of lessons and tasks.
All together, the observers identified and coded seventy-six lessons and teacher-assigned tasks. Thirty-eight percent of those were active, 35 percent were passive, 13 percent were constructive, and 12 percent were interactive. Observed student engagement with the lessons and tasks—meaning the mode students were in, versus the mode teachers intended—was coded separately. But there was a high degree of correlation—except that students were observed in passive mode more often than any other, especially in the hard-science classes and in the less advantaged schools. In other words, some lessons and tasks designed to be active, constructive, or interactive instead produced passive student engagement.
The researchers deem all of this to be problematic for numerous reasons—including the possibility that it shows teachers who believe that engaging with difficult material is beyond the capabilities of certain students and are choosing to teach “downward” to an entire classroom based on that belief. However, their concerns are predicated on several questionable convictions. Such as that ICAP’s interaction hierarchy is gospel and that students cannot fully and properly understand, say, physics without a specific (and much higher than observed) percentage of constructive and interactive engagement with the material. Their basis for concern is also dependent on the observation of a single lesson, which is not able to account for the flow of a semester- or year-long course, no matter how accurate their coding instrument may be. Many such courses, especially in the hard sciences, typically start with introduction of new and unfamiliar material via lecture and progress to independent and group activities (experiments, reports, and the like) designed to demonstrate and reinforce the new concepts. For all we know, the very next lesson from those teachers, not observed or coded, would have flipped the engagement mode timing entirely around with a whole-class discussion or an entire class period of hands-on lab work.
It’s not a bad idea for teachers to understand how much time they spend in each mode—the observational framework does seem adequate to that purpose—and to provide advice and techniques for them to keep students meaningfully engaged as much as possible. But this research does not yet go far enough to actually prove how much passivity is occurring or whether it’s a problem no matter what that percentage turns out to be. More observation, coding (AI, anyone?), and analysis are minimum requirements.
Editor’s note: This essay is an entry in Fordham’s 2023 Wonkathon, which asked contributors to answer this question: “How can we harness the power but mitigate the risks of artificial intelligence in our schools?” Learn more.
Having spent years in both university and school district administration roles, I’ve seen first hand how difficult it is to implement new, innovative ideas that have the potential to serve kids, teachers, and parents with fidelity. Artificial intelligence (AI) use in schools is going to be a complex effort that involves responsibility from various stakeholders, including vendors, school districts, teachers, students, and the broader educational community. It is going to require systems of accountability to be in place that are more agile, transparent, iterative, and responsive as the technology evolves and we learn more about its use cases, enjoy its benefits, and minimize its harms.
Here is my roadmap for how to ensure that, across a school system, anyone responsible for developing, supplying, procuring, implementing, and evaluating AI solutions is accountable for delivering what is best for students.
Start with the vendors
School systems have a lot of power through their procurement processes to ensure certain guardrails, parameters, and expectations are clearly articulated and adhered to by any vendors providing products or services. And while sometimes it may feel like procurement is where all exciting and innovative ideas go to die, a well developed RFP and procurement process that is anchored in a clear and coherent vision for what a district or school is looking for can be the first step in establishing strong expectations.
School systems should establish expectations from the outset by probing vendors to ensure their track record and core values are clear; there are clearly articulated and defined ethical considerations and data privacy policies in place; information about how AI algorithms work and what data is used to train, improve, and support the AI are transparent; and no ambiguity about how student data or information will be used. I’ve heard some call data “the new oil,” and districts have a legal and ethical obligation to protect students’ data and information.
Align use with developmental and pedagogical purposes
Beyond vendor selection, AI tools must continuously demonstrate and prove their efficacy and alignment to evidence-based, age-appropriate, and pedagogically relevant practices. The age of the students, their learning needs, and the specific educational context should all inform the choice and deployment of AI tools. Tools that are out of step with the age or learning objectives of the students will undermine the educational process and do not have a place in schools.
Some might argue that we’ve seen many tools, resources, materials, and technology used in classrooms that have proven little to no efficacy, alignment to evidence-based practice, or improved outcomes, so why are we over scrutinizing AI? The reality is, we should be scrutinizing and holding to account all of those other random things cluttering our schools and classrooms in the same way too.
Craft good policy
The days of no phones, no laptops, “no looking things up” are over. Policies that over-prescribe limits on the use of technology or uptake of new innovations are out of touch. Our school systems and classrooms should support, prepare, and incentivize educators and young people to tap into the technological resources they have at their disposal as they learn to critically think, read, understand and apply mathematics, and collaborate with their peers.
School districts must establish clear policies and guidelines that govern the use of AI within their systems to ensure classrooms are safe and productive learning environments. They must ensure AI policies and use are evaluated regularly and adjusted to reflect the evolving needs of students and advancements in technology. What we know today about AI will be drastically different five years from now, and policies should allow room for change, innovation, iteration, and learning while maintaining alignment to educational objectives and ethical standards.
Develop your people
Supporting teachers to understand and leverage the power of AI, integrate AI into their teaching methods, and adapt their pedagogy to accommodate these digital tools will be critical. AI can provide teachers with meaningful insights and identify areas where students need more support, which is desperately needed as we continue grappling with the effects of the pandemic on student learning.
But as AI tools advance to be able to surface needs and identify targeted supports, school leaders, districts, and policies should ensure that teachers’ professional judgment and triangulated data sources remain critical components in any intervention decisions. Humans must continue to make the final educational decisions that directly shape and impact students’ learning and continue to use professional judgment to ensure students are being served equitably and fairly.
AI tools must serve to augment, rather than replace, teachers, and should be used to enhance, not diminish, the role of teachers. Who doesn’t want to make the life of a teacher a little easier? Can we make teachers’ lives easier and, in the process, support them in personalizing learning to guide students towards mastery of grade-level content, critical thinking, and technological savvy? Sign me up!
Get to know and build skeptical trust in AI recommendations
Educators should also understand how AI systems make decisions and recommendations, particularly when it comes to recommendations about interventions or supports that target resources to specific groups of students. When users can’t grasp the basis of AI recommendations, the trust in these systems diminishes. Some call this the “black box,” where AI systems make decisions that seem arbitrary or inexplicable. AI tools should provide clear explanations of their reasoning, allowing users to have confidence in the recommendations they receive, and accountability policies should put in place checks and balances to ensure that no AI recommendations are implemented without scrutiny.
Involve students and parents
Parents and students should be at the table helping shape decisions for how AI is used at school and at home. They should be active participants in decision-making processes, informed about how AI is being used in the classroom, and encouraged to provide feedback and voice their concerns. Parents and students can become advocates for responsible AI use in schools and contribute to a safer and more effective learning environment if and when given the opportunity to do so. They can also play a critical role in ensuring accountability when things are not progressing positively or change is needed.
Pay attention to bias
At its core, accountability in AI use in schools must include a focus on values, ethics, and fairness. AI systems should be designed to be fair, unbiased, and free from discrimination. They should not perpetuate or exacerbate existing inequalities. Schools and districts must ensure AI tools are trained using representative, diverse, and vetted samples with safeguards against algorithmic bias, stereotyping, and the potential for discrimination in the use of AI. When an AI tool demonstrates bias, accountability policies should be in place to retrain and reprogram the tool to ensure it does not happen again.
Assess impact
No one fully understood how smartphones or social media would transform every aspect of our life in the span of fifteen years. AI is a dynamic field, and its impact on education is beyond what any of us could probably comprehend today. The only way we can keep up is by building strong guardrails and regularly assessing and evaluating the extent to which AI tools are enhancing educational outcomes. We must also constantly anticipate and respond to unintended consequences as they emerge. This should include information from academic assessments, surveys, and feedback from teachers and students. The data collected should be used to refine AI implementation strategies and inform policy decisions.
Editor’s note: This essay is an entry in Fordham’s 2023 Wonkathon, which asked contributors to answer this question: “How can we harness the power but mitigate the risks of artificial intelligence in our schools?” Learn more.
In that study, 582 kids (11 percent) were willing to use Khan as recommended (thirty minutes per week), while 4,766 refused. In fact, more kids in this study (748, or 14 percent) did zero minutes per week all year.
The 11 percent of “willing” kids had nice learning gains, estimated at 0.2 standard deviations. The low use kids (3,092) made no gains.
Imagine, we replicate this study in 2025, substituting AI-powered Khanmigo for OG Khan Academy. (You might further imagine some but not all bugs have been fixed.)
What do you bet will happen?
What percentage of kids will use Khanmigo as recommended, compared to the original experiment? That goes to motivation.
Will those kids learn the same, more, or less than the original experiment’s 0.2 SD gain?
Let’s return to this bet later.
3. What context do we need to ponder AI impact?
The first question: How good are 1-1 human tutors?
What you believe about AI’s ability to substitute for humans is dependent on whether you generally think human tutoring is great, good, OK, or bad.
There are four popular, competing answers about human tutor quality. We’d label them as:
“Bloom utopian,” “HDT evangelist,” “Mathematica pragmatist,” and “District superintendent pessimist.”
Let’s quickly examine.
1. There is a ludicrous study by the Bloom’s Taxonomy guy. Bloom once wrote that tutoring gets a 50th percentile kid to the 95th. That’s 2.0 standard deviations of gain. Bloggy takedowns of that study are here and here—and for my money, still too generous to Bloom. Yet this utopian study still gets a lot of play, from important people! The CZI and Gates Foundation “Moonshot” was framed using this research, for example. Yikes.
2. Back on Planet Earth, a top-cited recent Tutor Evangelist study is called The Impressive Effect of Tutoring on pre-K–12 Learning. They find “High Dosage Tutoring” moves a 50th percentile kid to the 62nd percentile. Sociologist Bo Paulle and I argued in 2020 that this estimate—while flattering to work Bo and I personally have done—overstate the real life terrain of human tutors in schools.
3. Gates Foundation generously hired Mathematica to measure more human tutoring programs. They did, in fact, find much lower effects on kids in their 2023 publications, like 0.12 SDs from Blueprint and 0.13 SDs from Air. Since these real-life effects were two-thirds lower than the Evangelist findings, I call this tribe the Mathematica Pragmatists.[3]
4. The District superintendent pessimist thinks: “But that’s not what happened with us. We bought tutoring and almost no kids were actually tutored!” The Pragmatists, Evangelists, and Utopians all respond: “You goofball. You didn’t actually buy High Dosage Tutoring, where each kid is supposed to sit down with a tutor at the following times each week. You bought a scam.” To which the Supe replies, “I don’t appreciate your critiques. We did the best we could.”
Anyway, that’s the wide range of perception about human tutoring quality.
What you think about AI’s promise depends in part how good you think the human substitute is. If you think human tutors have low impact, it’s hard to imagine how AI would be better. If you think human tutors have huge impact, then even diminished AI’s would be valuable with gains that are half or a quarter as big.
4. Human tutoring is really two things
Human tutoring is really two different interventions in one—a motivation intervention (like a personal trainer saying “c’mon, give me thirty more seconds of this plank!”) and an instructional intervention.
Remember, most in-school tutoring providers serve assigned kids—a mix of motivated kids and unmotivated ones, probably more of the latter.
Do we have any good data on private tutoring? You know, when Mom or Dad just hires a tutor?
Not really.[4] Understandably, private companies can’t easily ask half their potential customers to “not receive tutoring.”
Privately purchased tutors tend to have more motivated customers. For example, our daughters (Mike has one, Sean has two) are tutored by amazing women (Rashmi and Smriti). The results have been great, both to increase confidence and knowledge. As best Mike can estimate from MCAS and i-Ready scores over the past three years, our kid went from the 70th percentile to the 90th…putting her personal gain of 0.76 standard deviation gain somewhere between the Tutor Evangelists (0.37 SD) and the Bloom Utopians (2.0 SD).
Motivated students have more to gain from AI tutors.
By contrast, let’s empathize with the situation inside schools. The main driver of tutor impact is how well they deal (or not) with student motivation. Let’s loosely describe student motivation as red light, yellow light, green light.
Red light means a kid actively resisting tutoring: defiant, maybe head down on desk, perhaps literally wandering away from the desk, perhaps ignoring tutor and playing with phone, perhaps sitting arms crossed and staring daggers at tutor, perhaps video and audio off for an online tutor. This happens a lot!
Green light means a kid is willing to consistently try. The tutor assigns a problem, he tries; the tutor asks a question, he responds.
Yellow light is perhaps a kid who is up and down. Motivation may be low today based on his lousy weekend, high tomorrow after a great interaction with friends at lunchtime, and low again the next day because he failed a history test the period before, then back up because a kid he doesn’t like is absent.
A human tutor walks on eggshells with red and yellow light kids. If you push them hard to think, they may have an intellectual breakthrough (good), or may stop trying entirely (bad).
It’s hard to imagine how AI tutors will flourish in these types of situations.
5. Well, what about edtech interventions from the past twenty years, the various efforts to provide computer-based tutoring? After all, AI is supposed to be an improvement on that category.
Generally, these have generated lower gains than human tutors. That’s in part because ed tech is just an instructional intervention, not a motivational one.[5]
If we look at randomized trials of the likes of Zearn and Dreambox and similar products inside public schools, the most common finding is large numbers of students simply unwilling to use those products in the way they’re recommended, just like the Khan finding from Long Beach.
The average learning gains are in the range of 0.03 and 0.04 standard deviations. This moves a student from the 50th percentile to the 51st or 52nd. Those might be compared to the Mathematica pragmatists, finding 0.13 effects for human tutors.
6. Taken together, the data on human tutoring and computer tutoring allows us to estimate the “motivation” effect.
If we stick with the curmudgeonly estimate of human tutoring and the curmudgeonly estimate of computer software:
Human Tutors 2023 Mathematica (Motivation + Instruction): ~0.13 SD
Our hypothesis is that the main difference is motivation.
One can quibble with the studies chosen here. For example here’s a meta study that finds larger gains to software. The more rigorous the study, the more the effect size of both human and computer tutors declines—but the theme holds of human > computer.
7. So where does that leave us with AI?
AI is already often better at instruction than the incumbent edtech software. Already we see that Khanmigo helps a confused (yet motivated) student more than OG Khan Academy. AI will be great with green light, motivated kids.
Will AI help with red light kids?
Not that often, we predict. Yes, we know advocates imagine cuddly robots who “reach” these children. And there’s promise here for autistic kids in particular. But we’re generally skeptical.
Will AI flip some yellow light kids to green because its improved instruction generates some learning success, and that in turn leads to higher motivation? Yes!
So let’s return to the bet: If, in 2025, we replicate the Khan Academy Long Beach study….
What percentage of students use AI-powered Khan as recommended?
We’d bet 20 percent compared to the original 11 percent.
What learning gains would those 20 percent make?
We’d estimate 0.3 standard deviations, compared to the original 0.2 SDs.
To us, that is amazing. What a gain!
To those looking for an AI game-changer for students inside schools, sorry, it’s not quite that.
Will children—actually people of all ages—massively increase their consumption of AI tutoring for things they want to learn?
Yes!
Will AI manage to motivate the majority of schoolkids to grapple with all those subjects we policy wonks find important, but they typically resist?
No.
[1] Note: “motivated” here is shorthand for “motivated to learn the particular subject the school wants them to learn.” The same kid may be “unmotivated” to learn math and cricket and motivated to learn history and basketball.
[2] For purposes of this essay, let’s talk just about one-on-one tutoring (from humans or AI) and leave out classroom teaching. It simplifies things.
[3] Mathematica published some other studies we’re not mentioning, both that would reduce and raise this tutoring estimate. One was a zero result but there were study problems; another had a weird outcome variable instead of a normal standardized test like MAP or a state exam.
[4] There certainly are studies but they usually don’t compare well, for understandable inherit limitations, to what can be learned about tutoring in schools, randomized with the annual standardized test as the outcome of interest. We personally do hope to pioneer some research in this area.
[5] We concede that the software makers claim they do try to motivate kids—offering points, rewards, and badges. But given the RCT results, those inputs probably don’t move the needle that much with yellow light and red light kids. We don’t fault the creators. It’s hard to motivate!
