The bulk of commentary and school district policy relating to AI and education focuses almost exclusively on questions regarding cheating and whether it still makes sense to teach kids to write. But there are other risks of the tech and thorny questions looming on the horizon that are worryingly overlooked. Student privacy is perhaps chief among them.
The bulk of commentary and school district policy relating to AI and education focuses almost exclusively on questions regarding cheating. What does it mean for a student—or an education columnist, for that matter—if a chatbot can write a ten-page essay in a matter of seconds? How can teachers assign any homework and know for certain who or what is actually completing it? These are real questions with which the education sector must wrestle.
But there are other risks of AI and thorny questions looming on the horizon that are worryingly overlooked. Student privacy is perhaps chief among them. This concern became real to me when I read the following theoretical example from renowned education scholar John Hattie:
You are a sixteen-year-old girl on your way to school. Just before you get to the school gates, you waver and decide you would rather go meet your friends at the mall. Your Robo-Coach has access to the messages you exchanged with your friends the previous evening, it can sense your elevated heart rate as you stop to consider the alternative course of action, and it notes from your GPS location and lack of speed that you have stopped at an unusual place. From the combined data, it assesses an 87 percent probability that you will not attend school, and it interjects with a timely and personalized nudge (in a soothing versus commanding tone; male versus female voice; using an emotional versus factual plea, etc.) to push you through the school gates.
Chronic absenteeism is a real crisis in education right now. Research has found that regular alerts sent to parents about missing assignments and absences can improve pass rates, attendance, and retention. But this technology relies on human operators. Regular updates are time intensive—parent phone calls that I’ve made have lasted an hour—when teachers, administrators, and office assistants have other pressing concerns.
AI could supercharge this basic intervention without any extra resources or time demands on already overburdened school staff. Imagine if thousands of potential dropouts received such messages or parents got a simple text every time their kid missed an assignment, failed a test, skipped a class, or received a detention. Even if such messages only influenced a small percentage of cases, across millions of students, that would be a substantial nudge in the right direction.
However—and here is where student privacy enters the equation—in order for this actually to work, any such software requires access to vast sums of private, personal data: messages, family history, GPS location, previous attendance records, and biometric and basic identifying information. AI software is only as good as the data on which it is trained. An essential infrastructure for AI to function well is information. If electricity needs power plants, AI needs data stores.
Modern large language models (LLMs) like Chat GPT, for example, are distinctly powerful because they have essentially read the entirety of the Internet. With trillions of examples of text, it is functionally a high-powered, predictive language machine. What word is likely to come next based on the topic at hand and the words that preceded it?
This general language capacity can then be honed through more specific datasets. Ask ChatGPT to consume educational research in the What Works Clearinghouse, for example, and the chatbot can become an expert pedagogical advice-giver. The better the data, the better the output.
At the fringes of this conversation, we approach what once seemed like science fiction. Some schools in China already require students to wear biofeedback headbands that send information to teachers about who is paying attention, who is angry, who is daydreaming, or who is drooling on their desk. Pair that with the recent advances in brain imaging that have produced relatively accurate text from brainwaves alone, and soon our own private thoughts will not be so private.
There are obvious concerns about data breaches, as when hackers accessed an online test-proctoring platform and subsequently leaked private information about 444,000 students. Similarly, glitches in a popular program that lets teachers view and control student screens would allow hackers to gain access to students’ webcams and microphones.
But I’m more concerned about a more abstract threat: How does constant surveillance affect the way we think, speak, and behave? French philosopher Michel Foucault popularized the concept of the Panopticon, a theoretical, circular prison with a watch tower in the middle that shines bright lights into exposed cells. Prisoners feel a constant sense of surveillance and so act as if they’re always being watched—fostering a compelled self-regulation into “proper” behavior.
What happens when our students really are being constantly monitored? A report from the National Association of State Boards of Education suggests that students are less likely to feel safe enough for free expression, and that these security measures “interfere with the trust and cooperation” and cast “schools in a negative light in students’ eyes.”
What district policies do to address student privacy concerns do not even begin to reckon with such threats. Gesturing at FERPA and warning students not to enter private information into LLMs, as Oregon has done, is insufficient, but at least the state has begun thinking through policies and recommendations. The next phase should be to develop sensible systems and regulatory frameworks that could handle these kinds of difficulties.
Software companies can create the skeletons of the tools to sell to districts, which already maintain private records. Consent forms could allow access to plenty more data, but such forms should be clear and understandable, not the inscrutable tomes that everyone currently ignores. Regulation must compel companies to collect only necessary data, limit their ability to sell this information, and dictate not just parameters for data collection and retention, but also deletion. There must also be aggressive legal consequences for firms that mishandle such data and suffer breaches.
But questions still remain: Will data stores like those held by the National Center for Education Statistics become public utilities? Will we need “freedom of thought” court rulings and regulatory frameworks in the coming years? Will student “credit scores”—currently created and tracked by some districts using AI software—get passed along to police when they graduate?
Asking these questions makes me feel like I’m a bit of a crazy person. But these are not issues to face in some distant future of some science fiction novel. Many are here now, and waving away generative AI as just another forgettable edu-tech promise risks exposing students to all of the hazards of AI without harnessing its potential benefits.
When Emerson Elliott passed away the other day at eighty-nine, surrounded by his loving four-generation family, one might simply say adieu and thanks to a capable and dedicated career federal civil servant—which he surely was. But that’s not the whole story and we—education folks especially—are in his debt for so much more.
That certainly goes for me. Emerson and I interacted off and on for half a century. I met him in 1969 when I was a junior White House staffer with some responsibility for education within the domestic policy team and he—just fourteen years out of Albion College but already a twelve-year veteran of the Bureau of the Budget (now OMB)—was in charge of the federal education budget. (His boss at the time was the late Dick Nathan, another distinguished public servant with whom I had the honor of crossing paths on multiple occasions.)
I was a callow youth. Em knew what he was doing. And like senior OMB careerists were (and are) supposed to, he participated fully in working groups and policy powwows, offering his best judgment as to what might work best to accomplish some desirable end, how much it might cost, what’s feasible and what’s not, what would need to change, and so on. Professional and neutral, to be sure, but also informed, insightful, and wise.
We would re-engage in a big way a decade and a half later, but during that interim, Emerson moved to the old Department of Health, Education, and Welfare, then the new Department of Education (and also took a breather for two years at George Washington University). Here’s a bit from his resume:
November 1982 to 1984: Director, Issues Analysis Staff, Office of the Under Secretary, Department of Education
June 1981 to October 1982: Director, Planning and Evaluation Service, Department of Education
February 1981 to May 1981: Consultant to the Secretary on legislation and Department organization
May 1979 to January 1981: Director, HEW School Finance Project
March 1978 to April 1979: Director, Educational Staff Seminar, George Washington University
August 1977 to February 1978: Practitioner in Residence, Institute for Educational Leadership, George Washington University
August 1972 to July 1977: Deputy Director, National Institute of Education.
May 1972 to July 1972: Associate Deputy Commissioner for Management, Office of Education, HEW
As you can tell from the “director” titles, Emerson had emerged as a leader, not just a doer or staffer.
We had various small-scale dealings during that period, but were thrown back together in 1985 when I joined Bill Bennett at ED as assistant secretary, heading what was then called the Office of Education Research and Improvement (OERI), precursor of today’s Institute of Education Sciences. As we put together a team to run this section of the Department, we realized both that it needed a full-scale reorganization, and also that its statistics unit—the National Center for Education Statistics—was sick: underfunded, neglected, abused, sluggish, underproductive, and nowhere near what had been envisioned as much as 120 years earlier when Congress created the original “department of education” to “gather statistics and facts on the condition and progress of education in the United States and Territories.”
Two years after A Nation at Risk, the dearth of up-to-date, high-quality education data was a serious problem for the nation. Bennett knew it, I knew it, Em knew it—and an expert panel declared that either NCES needed a total overhaul or should be put out of its misery.
Emerson at this point assumed a dual role, first as major advisor and strategist on a big reorganization of the whole of OERI (which Congress eventually okayed), and also as the first modern “Commissioner of Education Statistics,” responsible for directing NCES. But it was more than directing. It was also rebuilding that weak entity into a semblance of today’s multi-dimensional federal education statistics agency.
With Secretary Bennett’s blessing (though to the consternation of the heads of OERI’s other four units), we directed pretty much every extra appropriations dollar we could get into the NCES budget. But that wouldn’t have solved the problem without Em’s solid, focused, diplomatic-yet-firm leadership of the place, which inevitably extended beyond the agency itself into deft if usually quiet dealings with colleagues, Congress, and innumerable keen-eyed (and often hungry) outside constituencies.
While doing all that, he also served as an indispensable member of the little brain trust (along with Mike Kirst and myself) that did the heavy lifting for the Alexander-James task force that reinvented NAEP to become the semi-independent and generally respected arbiter of “how good is good enough” in the achievement of American K–12 students and the honest rapporteur of their performance at the national and state levels (and now in a number of big cities, as well.)
I stayed at ED into 1988, but Emerson remained NCES Commissioner until 1995, when, still in his early sixties, he retired from the federal civil service, having accumulated by then pretty much every prize and honor available to top performers in that line of work.
Whereupon he embarked on what amounted to a second career of research, writing, consulting, and advising for a host of education-related organizations, particularly the teacher-ed accreditation body known as NCATE and its successor, the Council for the Accreditation of Teacher Preparation (CAEP).
But there was much more. To quote again from his resume (which goes only to 2012):
Since leaving the Federal Government in 1995, consulting and advisory memberships have included: the U. S. Department of Education, National Educational Research Policy and Priorities Board; National Academy of Sciences panels on Bureau of Transportation Statistics, a feasibility study for a Strategic Education Research Program, the Board on Comparative International Studies in Education (chair), and a workshop on national education indicators; the Advisory Committee on Research and Development for the College Board; the Science Resources Studies Advisory Committee of the National Science Foundation; the advisory panel for “Measuring Up” indicators published by the National Center for Public Policy and Higher Education; an evaluation of the Delta Cost Project for use by its Board to make long term strategic decisions; and education indicators for the State of the USA key national indicators project.
Most recently, Em and I reconnected in 2020 as I embarked on a book about NAEP’s history and current challenges and sought to consult his memory on various issues. Though well into his mid-eighties and soon headed toward what would become his final illness, he responded promptly, generously, and far beyond my expectations. He basically read every word of my manuscript, especially the history parts, amplified and corrected the facts, engaged respectfully on a number of touchy issues, and even used this occasion to rethink some longstanding points of tension between the “NAGB view” and the “NCES view” on key topics such as NAEP’s “achievement levels” (basic, proficient, advanced). It’s scarcely surprising that Emerson’s name appears on multiple pages of that volume, as well as in the acknowledgments.
I can’t claim that we were ever close friends. But I do assert, without fear of contradiction, that we were fortunate, as education folks and as Americans, to benefit from six decades of the highest-quality professionalism, conscientiousness, and near-perfect instinct for what would be best and work best and yet be doable from this remarkable public servant. May he rest in peace.
When former mayor Bill de Blasio promised to dismantle New York City’s gifted education programs, then-candidate Eric Adams laudably promised to save, reform and expand them. Since taking the helm, however, his actual policies and actions deserve scant praise—middling improvements paired with major regressions.
Earlier this year, for example, the Adams team expanded the number of sites for gifted education at the elementary level but scrapped the use of tests as a screening component—despite ample research supporting their use—and made school grades the sole determinant for inclusion. It then set the standards so low that two-thirds of students qualify for services, making the eligible range of achievement levels so wide that the programs are now “gifted” in name only: They can no longer feature the increased pacing and rigor that makes gifted education worthwhile.
Taken together, hizzoner’s changes have transformed gifted education in the lower grades into a glorified lottery based on the wrong measures, with students earning the best grades no likelier to win seats as those who barely make the cutoff.
As for high schools, the mayor did great work by creating three new selective schools that will open this fall—reducing needless scarcity. But that work is incomplete, as demand still far exceeds supply, forcing lotteries to operate here, as well.
To Adams’s credit, he clearly recognizes how much we risk if we’re cavalier with the education of America’s top-performing students. He understands that we need advanced learners to be highly educated to ensure our long-term competitiveness, security, and innovation. He also inherited a situation that was confusing, contentious, and headed towards abolition. Resurrecting it alone warrants applause—and may reverberate elsewhere.
Because of its size and cultural footprint, New York City models policy for countless other school systems, directs curriculum and textbook sellers, and influences education debates around the country.
And our country is in need of a stellar model in the realm of advanced learning. Years ago—immediately following Sputnik, for example—the U.S. took it seriously. But half a century of education policies have shoved it to the side, even to a place of skepticism and derision. Only two-thirds of districts offer any such programming, for instance, and in many places that do offer it, services are nothing more than meager supplements that are unlikely to make much of a difference in their students’ achievement, and are staffed with teachers who have not been trained to educate bright students.
In pursuit of a roadmap for districts across the land to do a better job of educating their advanced learners, a working group of diverse education experts—including one of us—convened last year to develop a report that summarizes the research on gifted education and recommends best practices. Its analysis suggests that New York City faces three big but solvable problems in this realm: the process for screening students is prone to bias and error, the programming in the city is needlessly all or nothing, and gifted services are far too scarce.
To solve problem one, the city should screen every single student in its schools using not only teacher-conferred grades, but also data from state standardized tests. Doing this in every grade for every child, instead of relying on a single age-based screening point that students must opt into, is an empirically-grounded method to identify intellectually gifted children from all strata of society.
For problems two and three, Adams should expand the number of seats at the elementary level, open more selective high schools, and mandate that every middle-school pupil in the city has access to advanced learning opportunities. In place of the city’s current dichotomous, you’re-in-or-you’re-out system where selective schools serve only a selective few, offer a “continuum of services”: separate schools or classrooms will be best for certain students, for example; in-class achievement grouping will suffice for others; and some should skip entire grades.
Mayor Adams is on the right track and deserves praise for rescuing advanced education from the abyss. They no doubt hope to eventually construct a gifted education system that serves every advanced student. But in its current form, the system falls short.
There is plentiful research suggesting that, among in-school factors, teachers consistently matter the most when it comes to student testing outcomes. On average, those with more experience and better prior preparation are likely to produce larger learning gains. Plus, teachers’ licensure status and their scores on licensure tests are positively linked to those gains. But this existing research is almost entirely focused on academic subjects such as math and English language arts (ELA). Enter a group of (mostly) CALDER researchers seeking to determine whether those patterns also hold in the realm of career and technical education (CTE).
Their data come from Massachusetts, which has led the way in building a robust CTE infrastructure and making an array of courses readily accessible to students in urban, suburban, and rural areas. The Bay State has notched some impressive outcomes for its CTE students, as well. Two-thirds of its CTE programs are known as Chapter 74–approved programs, which, among other things, makes them eligible to receive additional funds if they require their teachers to pass both a written and a performance subject matter test to get their preliminary Vocational Technical Education License. The performance test requires candidates to demonstrate skills in a hands-on setting where an evaluator assesses their tech skills and related knowledge. The programs are also required to have at least 70 percent of students eventually employed in the field of study.
The other one-third of CTE programs are dubbed “Perkins Only programs,” lacking (among other things) the funding benefit that comes from having test-based licensure. Thus, they make a handy comparison group for the researchers. Housed in both CTE-dedicated regional centers and within comprehensive high schools, Chapter 74 programs tend to enroll more White students and students with special needs than do Perkins Only programs, which tend to enroll more Black students. The study sample includes six cohorts of students with expected graduation rates from 2011 to 2016, comprising around 60,000 pupils across grades 9–12.
To address selection bias and sorting concerns, the study compares students in the same school and CTE cluster who take the same academic courses, have similar math and ELA test scores in eighth grade, and have similar demographics and school program participation. They also compare students within tracks so that they have similar course-taking patterns outside of CTE. That’s a lot of precision matching, but it’s still not a causal design.
The key finding is that, all else being equal, CTE teachers who received better scores on their subject performance test tend to have students with higher longer-term earnings than do CTE teachers who received lower scores. Specifically, a 1 standard deviation increase in teacher performance is associated with about a $900 increase in average expected earnings (approximately a 3.9 percent increase) for the teacher’s students five years after their expected graduation date, controlling for licensure test area and observable differences between students. The analysts benchmark that result against an analogous result in a seminal 2014 study that examined the impacts of teacher value-added on subsequent earnings (and other outcomes in adulthood). In that work, a 1 standard deviation increase in teacher value-added for students’ elementary teachers was associated with a 1.3 percent increase in annual earnings at age twenty-eight. Thus, CTE teacher value-added appears to function in similar ways.
The current study also suggests that the impacts are linear, such that increasing the performance of low scorers has the same impact as increasing the performance of the higher scorers. That relationship can be partly explained by the sorting that occurs into higher-paying occupations. But even then, higher scores on CTE licensure tests are independently predictive of higher earnings for students. What’s more, the scores on the performance-based, hands-on test are much more predictive of earnings than the multiple-choice, written test.
The analysts also find that teachers who are licensed in the program area in which they teach—as opposed to being licensed in a CTE cluster more generally—tend to have students with higher future earnings. But there is no relationship between CTE teacher experience (whether novices or veterans) and students’ four-year college enrollment overall. Oddly enough, there is a modest increase in the average long-run earnings of students who were five years out and had a novice CTE teacher in high school. The analysts hypothesize that, since these teachers have been more recently employed in their prior industry than veteran teachers, they may be better able to place students in a work-study position, provide current references, or make recommendations relevant in the immediate term.
Last, correlational design aside, the authors make this reasonable claim in their conclusion: “It is difficult to imagine how any licensure requirement like this could have positive impacts for students if the information collected during the licensure process did not contain any signal of teacher effectiveness.” Hear, hear! Yes, we need more rigorous research to say whether having CTE teachers with impressive, field-related licensure scores leads to high school students experiencing positive outcomes in adulthood. But for now, the signal is flashing green, so we must move ahead in identifying the most-talented teachers for this vital and growing education sector.
SOURCE: Bingjie Chen et al., “CTE teacher licensure and long-term student outcomes,” Education Finance and Policy (March 2023).
Data show that America’s current manufacturing workforce is aging and retiring as the sector is expanding exponentially and its demand for workers outpaces supply. Thus, a new report from the RAND Corporation looking at the career and technical education pathways designed to take postsecondary students directly from the classroom to the manufacturing floor is especially timely. It expands the research in this understudied area, focusing on efforts in Fordham’s home state of Ohio.
Data come from the Ohio Longitudinal Data Archive, drawing on information from the state’s higher education institutions and technical colleges, as well as from state unemployment insurance records. Student records include when each student first enrolled in a college or program; the duration of their enrollment; the program type, courses taken, and grades; whether and when the student earned a degree; and the type of degree earned. The technical college records are similar but include a course-based proxy for specific credentials earned. The data cover postsecondary enrollments and degrees earned between 2006 and 2019. The unemployment data cover 2007 through 2021, and all wage estimates are presented in 2021 dollars. To delineate “manufacturing-related” programs—the specific focus of their analysis—from other career and technical courses, the researchers use Classification of Instructional Programs (CIP) codes. As you might imagine, the list is heavy on engineering, construction, metalwork, and plastics courses.
In 2019, more than 30,000 students were enrolled in manufacturing-related postsecondary programs in Ohio. Approximately 1,600 were enrolled in technical colleges, 13,500 in associate degree programs at traditional two- and four-year institutions, and 13,800 in bachelor’s degree programs at those institutions. Interestingly, the previous decade showed a sharp decline in the number of students pursuing associate degrees and a steady rise in the number pursuing bachelor’s degrees. Technical college enrollment remained largely flat during that time. In terms of completion, approximately 7,700 students earned a manufacturing-related credential in 2019. Not surprisingly, given the enrollment trend, the largest share earned a bachelor’s degree. The number of completers at technical colleges was down against its high point a few years earlier. That total number also includes individuals who earned a manufacturing-related certificate independent of formal higher education settings (adult apprenticeships, formal upskilling by current employers, etc.), which nearly equaled the associate degree completers. In short, Ohio’s manufacturing employers should have been swimming in applications from skilled employees.
Data show that 82 percent of those who completed a technical college program were employed within a year of completion, compared to 84 percent of those who completed a certificate, 83 percent of those who completed an associate degree, and 65 percent of those who completed a bachelor’s degree. Those percentages remained fairly strong for up to five years post-completion, indicating that degrees and credentials are valuable to their recipients. However, a much smaller share of those credentialed individuals were primarily employed in the manufacturing sector. To wit: Just 38 percent of those who completed a technical college program were employed in Ohio manufacturing at one year post-completion, compared to 27 percent of those who completed a certificate, 25 percent of those who completed and associate degree, and 30 percent of those who completed a bachelor’s degree. Those percentages also remained similar after five years.
Where were they going instead? Construction, retail, technical services, administrative/support, and waste management/remediation services were the biggest recipients of manufacturing-credentialed workers in 2019, with a perhaps-predictable differentiation in professional versus service employment based on whether workers earned bachelor’s or sub-bachelor’s degrees or credentials. The attrition does not appear to be driven by industry pay gaps, as Ohio’s manufacturing industry pays higher wages than do other industries into which credential-earning workers opt. White males are more likely than females and non-white men to move directly from credential to employment, both in manufacturing and non-manufacturing jobs, and their wages are higher, as well.
The bottom line seems to be that even with a fairly clear pathway from the graduation stage to the factory floor, Ohio needs more newly-credentialed individuals to take that step, especially if a more-diverse manufacturing workforce is desired. Unfortunately, that is easier said than done. The report suggests geographic concerns (too few manufacturing jobs in the local areas where students are graduating) and skill mismatches (credentials earned are not the ones local employers need) as prime areas where gaps could be addressed. It is to be hoped that as the Silicon Heartland takes root and expands outward from central Ohio, the new and growing manufacturing presence here will bring on a concomitant effort to reach into the education world and help shape existing training and credentialing programs to their needs, to build new ones as needed, to welcome nontraditional employees, and bridge any gaps or mismatches between school and work.
SOURCE: Lisa Abraham, Christine Mulhern, and Lucas Greer, “Strengthening the Manufacturing Workforce in Ohio,” RAND Corporation (August 2023).
Cheers
Jeers
