Magic Data and the Dumbing Down of America’s Teacher Shortage

18 min readMar 13, 2024

This “knowledge” puffs up. — 1 Corinthians 8:1[1]

Writing for The Heritage Foundation,[2] Lindsey M. Burke, PhD is convinced there’s a “Phantom Teacher Shortage.” According to Burke:

[P]ublic schools have been on a hiring spree for decades. … Nationwide, there were 5% more teachers in America’s public schools during the 2020–21 school year than during the 2017–18 school year. … By 2021 … 3.2 million public school teachers [were] educating about 55.5 million students, shrinking the student-teacher ratio nationally to just 15 to 1. … This staffing surge has been a deliberate decision made by the education establishment.

So Burke’s assertion is that public schools have been:

· on a “hiring spree” (which can be defined as “a sustained, energetic period of hiring a lot of people”)
· for “decades” (the plural “decades” can mean nothing less than 20 years)[3]
· and it’s a “deliberate decision made by the education establishment.”

The Data

An Increase Yes, But ….

The data shows that yes, there were “5% more teachers in America’s public schools” from 2017–18 to 2020–21. What Burke leaves out is that in 2005, the student-teacher ratio nationally was 15.6 to 1. Then, for nearly 10 years, it jumped to (or above) 16 to 1, only to fall back to 15.4 to 1 in 2021 (in 2023, that number was still 15.4 to 1).

So for nearly a decade (2010–2018), teacher staffing was negatively impacted (because of reduced hirings or more students or both), dropping to a level that was closer to 1999 (a student-teacher ratio of 16.1 to 1). With the student-teacher ratio increasing (and holding) for nearly ten years, it’s hard to find a “hiring spree” that’s lasted for “decades.” More importantly, one can see that hiring “5% more teachers in America’s public schools” might well be a course correction, returning today’s student-teacher ratio (15.4 to 1) to levels not seen in nearly 20 years.

There’s also the question that Burke fails to answer: who exactly is a “teacher” in this “5% more” hiring spree? The data is based on “full-time and part-time traditional public school teachers,” meaning “5% more” doesn’t entirely mean “full-time” more (or to what extent). Moreover, the data itself doesn’t fully define “teacher,” meaning “5% more” might be a combination of full-time teachers, part-time teachers, classroom teachers, virtual teachers, and uncertified teachers. Thus, the question: who’s part of and who’s missing from this alleged “hiring spree”?

The Magic Ratio

Then there’s the optical illusion of “shrinking the student-teacher ratio nationally to just 15 to 1.” Here, Burke implies that all class sizes (“nationally”) are at a non-worrisome (“just”)[4] 15 to 1 ratio. However — and Burke neglects to acknowledge this — 15 to 1 represents both my sixteen-year-old daughter’s English class (25 to 1) and her screen writing class (5 to 1).

In his book Naked Statistics, Charles Wheelan explains why Burke might provide the average/national ratio and simply stop there:

The mean is a simple average: the sum of the observations divided by the number of observations (The mean of 3, 4, 5, 6, and 102 is 24). … Now, the clever reader will see that there is sizeable difference between 24 and 5. If, for some reason, I would like to describe this group of numbers in a way that makes it look big, I will focus on the mean. [emphasis added]

Using Burke’s logic, one can argue that overpopulation is a “phantom” issue because the “average population density of the U.S. is [just] 94 people per square mile.” This hides the fact that Los Angeles’ population density is 7.476 people per square mile … the densest major urban area in America. Burke’s ratio of “just” 15 to 1 hides the fact that many classes “greatly exceed” 15 to 1. And Burke’s not the first to try this; teachers have been fighting data twisting for years:

Diana Gross, a ninth-grade science teacher, told the Norman-area [Oklahoma] legislators that the formula for determining student-teacher ratios is misleading. Gross said she can’t teach students how to think in a hands-on science class when class size reaches 30 to 40 pupils. Her smallest class is 25 students, although the official school ratio of 16-to-1 gives the impression that class sizes are smaller, she said. “Most core curriculum classes[5] have more than 30 students to a class” on her teaching level, Gross said. Gross said she hasn’t seen a class size fitting the official ratio, “unless it’s a special education class.”

Note that the official ratio mentioned is “16 to 1,” equal to the 2018 national average. Burke’s assertion of a “hiring spree” is difficult to see in Norman, Oklahoma … considering the aforementioned article was written in 1989. This means Norman Public Schools’ student-teacher ratio has remained stagnant for nearly 30 years (compared to today’s national average). In fact, for the 2022–2023 school year, Norman’s ratio sat at 16.4, meaning the “shrinking” Burke touts never happened in the eighth-largest school district in Oklahoma.

By citing the national ratio and stopping there, Burke acts as if a teacher shortage (assuming there is one … I’ll get to this later) can be disproven with “just” a ratio … a data point that seems to support “shrinking” and “hiring sprees” and “staffing surges.” The truth is, the data is much more complex than Burke lets on:

· Some states are well above the national average of 15 to 1, while some are well below. For example, in 2021, Arizona, California, and Utah were well above the average at 22 to 1, while Illinois, Nebraska, and West Virginia were lower at 13 to 1.
· In 2021, the pupil/teacher ratio in rural fringe schools was 15.5; in rural distant schools, 13.7; in rural remote schools, 12.2. On the other hand, Los Angeles Unified School District has a ratio of 20 to 1, Miami-Dade County Public Schools is 19 to 1, and Houston Independent School District is 18 to 1.
· According to the Institute of Education Sciences (the source of Burke’s 15 to 1 ratio), the “pupil/teacher ratio includes teachers for students with disabilities [7.3 million students/15% of U.S. enrollment] and other special teachers, while these teachers are generally excluded from class size calculations.” Including “students with disabilities” — often taught in a self-contained classroom of 5–10 students — deceptively pushes the national average downward.

The Conspiracy

According to Burke, the phantom teacher shortage is a long con that’s been perpetuated by the teacher unions:

This staffing surge has been a deliberate decision made by the education establishment. … This choice made by teacher unions — to prioritize hiring more staff (teaching and non-teaching staff) over higher salaries — equates to more dues-paying union members, which is, of course, good for the unions’ bottom line.

One has to wonder why the union — if in need of money — wouldn’t simply raise dues rather than masterminding a hiring spree that — for “decades” — has fooled the media, fooled parents, fooled school boards, fooled administrators, fooled researchers, and fooled legislatures.

Regardless, Burke cites nothing to support her assertion that teacher hiring has become a cash cow for unions.[6] Moreover, she neglects to mention that unions are shrinking DESPITE the alleged “hiring spree” that’s been going on for “decades”:[7]

· From 1999 to 2016, teacher-union membership dropped nationally from 79.1% to 69.9%.
· From 2022 to 2023, membership decline in the National Education Association (the NEA is the nation’s largest professional employee organization) jumped from 7.3% to 8.1%.
· As reported in 2022, the Illinois Federation of Teachers lost 17,975 people paying dues or fees to the union, a nearly 18% decline in membership.
· In Minnesota, teacher membership dropped significantly, decreasing by 3.1% in a single year. This marks the third consecutive year that the state teachers’ union lost teacher members.
· Michigan’s union is down to 78,817 members — a loss of more than 1,000 since 2022. The union has shrunk by more than 38,000 members since 2013.
· In California, new hires are increasingly choosing not to join the union. As of the end of September 2023, 36,000 eligible public education employees chose not to join. That’s up from 18,000 in 2019.

So with this decades-long hiring spree, what are the spoils of war for America’s largest teacher union — the National Education Association (NEA)? According to ProPublica, the NEA’s main revenue (roughly 95% or above) comes from “Program Services’ revenue, which is defined as “funds received by an organization in exchange for providing the services for which it received tax-exemption (e.g., tuition, fees, or admissions).” For the NEA, this is what their main source of revenue looks like over the last 5 years:

· $374,451,660 (2022)
· $376,378,044 (2021)
· $375,030,303 (2020)
· $368,394,888 (2019)
· $373,762,760 (2018)

So with the help of Burke’s alleged “hiring spree” and the unions’ long-con money grab, 2022 produced a whopping 0.18% increase in NEA revenue compared to 2018 and a -0.51% decrease compared to 2021.

The Slight of Hand

Hidden within Burke’s argument is a “let’s sweep it under the rug” fact … “just” 28 words (buried among 800 other words) that get glossed over quickly:

In some school districts — mainly in the South and rural areas — and in some subjects (math, special education, foreign languages, chief among them), schools have trouble filling positions.

That’s it. Begrudgingly, Burke acknowledges that somewhere out there, a shortage exists. But then she quickly adds, “[T]his problem is not acute.”[8] However, it’s difficult to determine what’s “acute” or not in “just” 28 words. To the following schools, “trouble filling positions” seems more “acute” than Burke wants to admit:

· schools in the “South” (Burke’s quick mention lends itself to insignificance). According to a 2022 study from Brown University, there were approximately 36,500 vacant teacher positions across the U.S., with the southern region accounting for the highest portion, 22,600 vacancies, which was almost three times higher than the Midwest.
· schools in “rural areas”[9] (again, glossing over it implies “nothing to worry about”). More than 9.3 million students go to public schools in rural areas, more than the combined total of the nation’s 85 largest school districts. Teachers were over twice as likely to move out of rural schools and to urban or suburban schools as they were to move from urban or suburban schools to rural schools. The result is an annual reshuffling of a large portion of the rural teaching force, with a net loss to rural schools and a net gain to urban and suburban schools.
· schools in “some” districts (Burke makes little effort to define which districts (besides the South and rural areas) constitute “some school districts”). “Some school districts” include the San Francisco Unified School District (neither in the South nor rural). The 49,000-student SFUSD began this year with more teacher vacancies than Los Angeles Unified, which serves 480,000 more students. SFUSD students started this year with at least a quarter of its classroom vacancies unfilled.
· schools with higher shares of students of color. One in every eight schools (13%) with 75% or more students of color had teacher vacancies in excess of 10% of total teaching staff, versus 7% in schools where students of color made up less than 25% of students.
· schools in high-poverty areas. Fifteen percent of schools in high-poverty neighborhoods had teacher vacancies of 10% or higher, compared with 8% in low-poverty neighborhoods.

And what about this acknowledgment from Burke?

… and in some subjects (math, special education, foreign languages, chief among them), schools have trouble filling positions.

“Chief among them” implies that something’s missing … i.e., the lesser important subjects that didn’t (for whatever reason) make Burke’s list. In terms of teacher staffing, here’s what “chief among them” leaves out:

Shortages are most acute in some specialties, particularly special education (45% of schools reporting vacant teaching positions), mathematics (16%), English or language arts (13%), English learner education (13%), and physical sciences (10%) …. [A]lmost one-third (31%) of schools … [reported] vacancies for “general elementary” teachers and one-fifth (20%) of schools … [reported] vacancies for substitute teachers ….

The Experts

My wife — a 25-year veteran of public-school teaching[10] and a 2022 Site Teacher of the Year — laughed when I read Burke’s assertions. She laughed harder when I mentioned Burke’s declaration that the “shrinking” student-teacher ratio nationally was “just” 15 to 1. She laughed because in the 20+ years at her current school (i.e., “decades”), her class sizes have consistently surpassed 25 students. An important thing to note … her school’s student-teacher is listed as “just” 17:1.

One would think that my wife — and the nearly 4 million teachers in America — would be our best experts on the alleged teacher shortage. Not Burke, who as far as I can tell has no long-term experience in public school teaching. That doesn’t mean Burke can’t speak into this, but at the very least, we should question all “experts” … we shouldn’t be blinded by their doctorates, their publications, their foundations, and their data. As Nicholas Taleb, author of The Black Swan, has said:

The problem with experts is that they do not know what they do not know.

Burke can repeat numbers and data, but she doesn’t live — and doesn’t report on — the daily machinations of a public-school classroom:

From Fox News: “I think the biggest issue we’re having right now is there’s a national teacher shortage. Three-quarters of U.S. states now report that they are short on teachers. We have teachers leaving the profession in droves and I think that puts a lot of stress on the teachers that are in the classroom teaching. They’re not as well staffed, classes are overcrowded, and that’s a big stressor on teachers today.” — Teacher Brook Ooten
From the Huffington Post: “We’re hitting a crisis point; the broken system is breaking teachers faster than they can be replaced. … After last year, I had to walk away. Despite the unprecedented strain caused by the pandemic, for so many teachers, there has been no abatement of professional development, evaluation, or pleas to sub for other teachers from district leaders who choose to gaslight teachers with toxic positivity rather than address their concerns.” — Former teacher Katie Niemczyk

Of course, these quotes — taken directly from teachers — don’t represent the universal experience of America’s teachers, nor do they negate the joy that many teachers get from teaching. But first-hand observations — what happens in a classroom, what teachers directly experience — are often ignored by academics, number crunchers, and ratio magicians.

But What If ….

Earlier, I mentioned “teacher shortage,” then followed up with “assuming there is one.” I’m willing to listen to Burke’s argument that there’s a “phantom teacher shortage,”[11] but I can’t listen to half-baked numbers, one-dimensional data, and shallow explanations. Outside Burke’s box (or outside her bias), “teacher shortage” isn’t s a universal term … it means something completely different in Nevada (43.65 teachers per 1,000 students) vs. Vermont (97.6 teachers per 1,000 students). This means (as Burke implies) that not every classroom, district, or subject represents a teacher shortage; some data will show a surplus or, at most, “just” enough teachers to meet the needs of their students.

But data of one thing doesn’t automatically negate the data of something else (though it may hide that something else). As British economist Ronald Coase said, “Torture the data, and it will confess to anything.” Any article titled “The Phantom Teacher Shortage” tortures the data, though, admittedly, not every expert buys into this assertion. Earlier, I said that when it comes to experts, we need to hear more, believe more, and talk more to our teachers. Despite the overwhelming and opposite outcry from educators, there is one teacher (“expert”) — Peter Greene — who agrees with Burke … that “There Is No Teacher Shortage”:

It certainly feels as if teaching conditions are worse. The heated rhetoric is getting hotter, from attempts to micro-manage what teachers can teach. Pay remains stagnant. Respect seems as if it’s at an all time low. Every teacher knows a teacher who left the profession ahead of schedule, or a promising prospective teacher who chose not to enter the field at all.
The trouble with teacher shortage rhetoric is that it mislocates the problem. If we argue that all of the nuggets have been pulled from the mine, we don’t have to consider the possibility that there’s plenty of rich vein left, but we can’t mine it with a plastic spork.
There is no teacher shortage. There’s a teacher recruitment and retention problem. [emphasis added] There’s a “making the job attractive enough to draw in the people we want” problem. There is a problem that requires a careful, thoughtful diagnosis. There are policy and political leaders who see the current situation as an opportunity to be exploited rather than a problem to be solved. Those are not the voices we should be listening to right now.

[1] Or put another way … “The greatest obstacle to discovery is not ignorance — it is the illusion of knowledge.” — Historian Daniel J. Boorstin. Or as I like to say it, “A little data lays down a lotta fog.”

[2] As a conservative-leaning Christian, I agree with many things from The Heritage Foundation. However, that doesn’t mean I agree blindly. As readers, we should question powerful voices like The Heritage Foundation ($106M in reported revenue/2022), the same way teachers should question the powerful voices of the National Education Association ($378M in reported revenue/2022). As Jesus said in John 20:29, “Because you have seen Me, you have believed. Blessed are those who have not seen Me but still have believed.” Nowhere does Jesus command us to believe blindly.

[3] I’m not here to debate Burke’s assertion that hirings have increased. From the data, I’m willing to concede that yes, there have been more hirings (overall) in the education sector. What bothers me is the taking of microscopic, vacuum-packed, one-sided data and offering it up as the gospel (a not uncommon blight in modern writing and reporting).

[4] One might argue that “just” is just a word, but words matter … even small ones. A virologist who says “It’s the Ebola virus” means something different than the one who says “It’s just the Ebola virus.”

[5] Another thing that’s important to consider … the expansion of specialty classes beyond core curriculum classes. Typically, we think of student-teacher ratios (and their rise/fall) in terms of increasing/decreasing students or increasing/decreasing teachers. But there’s a third way to manipulate the ratio that Burke doesn’t consider … it’s the addition of more classes and/or specialties to a curriculum (something that’s in demand from students, parents, and policy makers).

“More classes to choose from” is a modern trend, one that’s not reflected by the basic course load of the ’60s and 70s. Today, students have more classes — beyond math, science, and English — to choose from, and many of these specialty classes (with individual teachers covering multiple subjects and specialties) have much smaller student-teacher ratios.

For example, at my daughter’s school (these are all real classes), “Algebra I” might have a student-teacher ratio of 25 to 1. But specialized classes might have ratios that look something like this … “Differential Equations” 15 to 1, “Chinese V” 10 to 1, “Advanced Placement 3-D Art & Design” 5 to 1. Greater selection and specialization creates more work for limited teachers (who often teach multiple subjects) but smaller classes compared to the core/basic subjects. In the example I just gave, Algebra I’s class size is hidden by the smaller specialty classes, which together represent an average student-teacher ratio to “just” 14 to 1.

[6] This is where I’ll be accused of being a union “sympathizer” even though 1) I’ve never been part of a union and 2) I haven’t seen the benefits of union membership in my home state of Oklahoma. According to the NEA (yes, it’s union data), Oklahoma’s average starting teacher salary ranks #42 in the U.S. at $38K, which is about $10K below the minimum living wage.

[7] One can imagine Burke’s counter argument … that because unions are losing members, this supports the idea of a responsive “hiring spree.” But this immediately kills at least one of Burke’s two main arguments. If the membership loss is because teachers are abandoning the unions, then this undermines Burke’s assertion that there are “more dues-paying union members.” “Hiring spree” gains are being offset (to some degree) by membership losses. On the other hand, if the membership loss is because teachers are abandoning the profession, then this undermines Burke’s assertion of a “phantom” teacher shortage. Here, a shrinking union membership signals more teachers leaving the profession, resulting in a need for new hires.

[8] If you want to make something “not acute,” the thing to do is generalize the data or narrow its focus or offer surface level analysis. For example, I can easily prove that the physician shortage (if there is one) is “not acute.” There might be a shortage in “some” places and maybe in “some” specialties, but “for decades,” the “physician workforce has steadily grown faster than the U.S. population.” (this over a 30-year period as reported in 2013). And as reported by the Harvard Business Review:

Some simple math would suggest that we should have more than enough primary care physicians. By 2025, the Department of Health and Human Services estimates we should have 190,000 non-pediatric PCPs. Assuming normal patient panels (about 2,000), that suggests the capacity to care for 380 million people (vs. a U.S. population estimate of 272 million).

So looking backward, the surface-level data proves years and years of physician growth; looking forward, the surface-level data estimates plenty of physicians in the U.S. So “not acute” despite my use of the word “some.” However, hidden within “some” are greater specifics, more data, and deeper truths:

From 2010 to 2020, the number of general pediatricians increased by 6.0% (53 600 vs 56 800 pediatricians), but fewer of them worked in rural counties (3000 vs 2900 pediatricians). Similarly, the number of FMPs [family medicine physicians] increased by 14.1% (79 400 vs 90 600 FMPs) with fewer working in rural counties (13 000 vs 12 000 FMPs). … This cross-sectional study found that although numbers of general pediatricians and FMPs increased nationwide from 2010 to 2020, fewer practiced in rural counties, likely re-entrenching rural-urban disparities in pediatric outcomes. Coupled with the declining percentage of FMPs treating children, the pediatric primary care workforce deteriorated in rural areas. [a/k/a “a shortage”]

And this:

The number of primary care physicians per 10,000 residents is generally higher in much of the Northeast, along the West Coast, in Hawaii, and parts of the mountainous West and upper Midwest. The availability of primary care physicians per capita is generally lower in much of the Great Plains — especially the Southern Great Plains — and the Lower Mississippi Delta and Southeast. [a/k/a “a shortage”]

[9] I have a friend who’s 76 years old, and his wife — an English teacher — took a teaching job in rural Midway, Oklahoma (239 students in grades PK, K-12 with a student-teacher ratio of 17 to 1). She took it because the school was having difficulty filling the position. At 70 years old, she drives nearly two hours (total) — every weekday — from Tulsa to Midway and back.

[10] My wife has taught at two low-income schools and at one of the largest and best public schools in Oklahoma. In that time, she has taught second grade, third grade, fourth grade, and STEM.

[11] “The Phantom Teacher Shortage” makes for a more compelling (and scandalous) title than “The Kinda Teacher Shortage,” which is really what the data suggests. “Shortage” isn’t a universal thing, but it’s out there in different places, at different levels, and for different people — “people” being thousands of teachers, students, and parents whose life experience contradicts Burke’s so-called “phantom.” Regardless, Burke’s not alone in her observation; this phantom has been suggested before. In 2022, the Hechinger Report reported “Researchers say cries of teacher shortages are overblown.” This headline is supported by data from RAND, a nonprofit research organization:

“We asked schools what shortages they expect for the 22–23 school year and they did not anticipate a huge shortage,” said [RAND researcher Heather] Schwartz. Three-quarters of the districts said they expect a shortage, but most of them, 58 percent, said it would be a small shortage. Only 17 percent of districts anticipated a large shortage of teachers.
Schwartz says her biggest worry isn’t current teacher shortages, but teacher surpluses when pandemic funds run out after 2024. School budgets will be further squeezed from falling U.S. birth rates because funding is tied to student enrollment. Schools are likely to lay off many educators in the years ahead.

But as is the case (almost always), there’s more to the story, more complexity, and more data than this. Because RAND has also reported:

In addition to day-to-day coverage issues, most principals reported that teacher vacancies were on the rise. Most of these principals believed that vacancies had grown more difficult to fill than in the prior school year, largely because of declining applicant counts. [emphasis added] … Most principals — both those with and without teaching vacancies in their schools — expressed that their hiring efforts had been impeded by low applicant counts, low compensation, and underqualified candidates. — “Principal Perspectives on School Staffing Struggles” | May 3, 2023