Privacy-friendly analytics help us see which jobs people explore. They stay off unless you agree.Analytics off unless you agree. Details

Read the record
Back to the river
Skip to sources
Time Machine

Math professors

Official title Mathematical Science Teachers, Postsecondary (SOC 25-1022)

Also called Mathematics professor, Math lecturer, Mathematics lecturer, Math faculty, Postsecondary math instructor

In brief

In May 2025, BLS counted 47,670 US math professors, with median pay of $79,940 a year. Of the BLS counts since 2021, the highest was 48,820, in May 2024, close to the May 2025 count. The earliest figure here is an estimate of 1,200 from 1870, not measured the same way as today's count. The last full technology era for math professors: Online homework systems and adaptive learning, 2005 to 2020. Online homework systems reduced the grading workload for large introductory courses, but also enabled departments to increase class sizes without proportional faculty additions.

What the sources say

Official outlook

BLS projects employment to grow 1.7% from 2025 to 2035.

BLS Employment Projections 2025-35

AI exposure studies

Data as of BLS OEWS May 2025; BLS Employment Projections 2025-35. BLS publishes these counts as yearly survey snapshots and cautions against reading them as a precise trend.

Last updated .

Cite this page
APA

Future History. (2026, October 11). Math professors: jobs and pay since 1870 [Data set; data as of BLS OEWS May 2025; BLS Employment Projections 2025-35]. https://futurehistory.earth/timeline/25-1022

Chicago

Future History. "Math professors: jobs and pay since 1870." Data as of BLS OEWS May 2025; BLS Employment Projections 2025-35. Last modified October 11, 2026. https://futurehistory.earth/timeline/25-1022.

BibTeX

@misc{futurehistory_25_1022, author = {{Future History}}, title = {Math professors: jobs and pay since 1870}, year = {2026}, month = oct, howpublished = {\url{https://futurehistory.earth/timeline/25-1022}}, note = {Last updated 2026-10-11. Data as of BLS OEWS May 2025; BLS Employment Projections 2025-35}, }

The access date is today's date on your device.

drag to travel through time
17501775180018251850187519001925195019752000now
2026
Known then as Professor of Mathematics and Natural Philosophy (colonial and early national era)
Known then as Professor of Mathematics (antebellum college era; often sole math faculty at a given institution)
Known then as Professor / Instructor of Mathematics (research university era; departmental expansion)
Known then as Mathematics Professor / Instructor (postwar expansion; GI Bill and Sputnik eras)
Known then as Mathematics Faculty / Postsecondary Mathematics Instructor (adjunctification begins)
Known then as Mathematical Science Teachers, Postsecondary (BLS SOC 25-1022)
Known today as Mathematical Science Teachers, Postsecondary (BLS SOC 25-1022)
Projected to still be called Mathematical Science Teachers, Postsecondary (BLS SOC 25-1022)
Data from the May 2025 survey
US Employment
48K
Est.

Stated in source: BLS OEWS

Official statistic
Report an error
Median Annual Wage
$79,940
Est.

≈ $77,890 in 2024 prices (derived by us, CPI-adjusted)

Official statistic
Report an error
Beat · 1727

Harvard College establishes the Hollis Professorship of Mathematics and Natural Philosophy, the first endowed chair in mathematics in North America. The chair is funded by Thomas Hollis of London, who wanted it modeled after Oxford and Cambridge professorships. Isaac Greenwood is appointed the first holder. Before this appointment, mathematics instruction at Harvard was handled by rotating tutors without specialized training in the subject. The Hollis chair marks the moment mathematics teaching in America becomes a distinct professional identity rather than a duty assigned to whoever can manage it.

Beat · 1825

Harvard becomes the first American college to offer calculus as a course, in 1825, available to seniors as an elective alongside chemistry. Union College follows with differential and integral calculus for seniors in 1828. The introduction of calculus as a university course creates demand for faculty with genuine mathematical training, not merely the ability to drill students in Euclidean geometry and arithmetic. By 1900, calculus has become a standard part of the mathematics curriculum at most American colleges.

Beat · 1862

Yale awards the first US PhD in mathematics to John Hunter Worrall. The appearance of doctoral mathematics education marks the beginning of the research-trained faculty supply: before 1862, American colleges hired mathematics instructors who had trained primarily in Europe (especially Germany) or had no graduate training at all. Within a decade, Johns Hopkins (1876) would formalize the PhD as the standard qualification for university mathematics faculty.

Beat · 1876

Johns Hopkins University opens as America's first research university, explicitly modeled on the German research university. J. J. Sylvester is hired as the first mathematics professor, establishing the principle that a mathematics professor must both teach and conduct original research. The Hopkins research seminar model (weekly meetings where graduate students and faculty discussed live research problems together) transforms what American mathematics faculty are expected to do. Cornell, Harvard, Chicago, and other major universities adopt the model within a decade.

Beat · 1944

The Servicemen's Readjustment Act of 1944 (GI Bill) passes, providing tuition, fees, and living allowances for veterans attending college. By 1947, 49% of all enrolled college students in America are veterans. Campuses that had been half-empty during the war are suddenly overflowing, and colleges scramble to hire faculty in every discipline. Mathematics departments expand rapidly to staff introductory calculus and precalculus courses required for the engineering and science programs veterans are enrolling in at unprecedented rates. The GI Bill is the single largest driver of mathematics faculty hiring in American history to that point.

Beat · 1957

The Soviet Union launches Sputnik 1 on October 4, 1957. Within 12 months, the US passes the National Defense Education Act (1958), which dramatically increases federal funding for science and mathematics education at all levels. NSF funding for mathematics programs increases by approximately 300% between 1957 and 1960. The SMSG (School Mathematics Study Group), funded by NSF and chaired by Edward Begle of Yale, assembles university mathematics professors to rewrite mathematics curricula. The resulting "New Math" restructures how calculus and abstract algebra are taught in American colleges through the 1960s and 1970s, and requires mass faculty retraining via NSF summer institutes.

Beat · 1988

Wolfram Mathematica 1.0 releases on June 23, 1988. Stephen Wolfram describes the moment: "There was a wow that one could routinely do integrals symbolically by computer." For mathematics faculty, Mathematica is the first software that can solve the problems on a typical undergraduate exam, a fact with immediate implications for how exams should be designed and what a mathematics course is supposed to teach. The debate between CAS-active and CAS-neutral instruction begins and occupies mathematics education research for the next two decades.

Beat · 2024

OpenAI's o3 model scores 96.7% on AIME 2024 (competition mathematics problems specifically designed to be beyond routine algorithmic solution) and 87.7% on AMC. Google DeepMind's AlphaGeometry has already demonstrated silver-medal-level performance on International Mathematical Olympiad geometry (Nature, January 2024). DeepSeek-Math has shown LLMs can produce competition-level solutions. These milestones arrive in the same year that Wolfram Alpha Pro and ChatGPT are being used by students to solve virtually all standard undergraduate calculus and linear algebra problems on demand. The MAA formally begins documenting the curricular redesign this requires: calculus courses across the US are being rebuilt around proof construction, AI-output critique, and oral mathematical defense rather than computation.

How to read this number

OEWS is a point-in-time survey snapshot, not a continuous time series; BLS advises against using it for year-over-year trend comparison.

Each dot is a cited figure over time; the dotted line only links them (values between aren't measured). Hollow dots are estimates.

Official statistic: Copied by a script from a statistics agency table or a published study's table, not retyped by hand. BLS figures and the forecast and exposure tables are also re-checked against that table on every data run. Researched context: Researched from the cited source rather than copied from it by a script, so it is not machine-checked yet. Spot a mistake? Each figure has a “Report an error” link.

Cited forecasts · present → 2035

Will AI replace math professors? What the research says

Each figure is stated by the named source; we publish no forecasts of our own. Where sources disagree, we show the disagreement rather than an average.

Each figure is stated by the named source; we publish no forecasts of our own. Where sources disagree, the spread is the honest signal.

What published forecasts say
Forecast change in the number of people doing this work, as published by each named source. Bars share one scale on every role, centred on zero and reaching the end at ±30%. Grey marks a small change, 3% or less either way.
BLS Employment Projections 2025-35
→ 2035
+1.7%
BLS National Employment Matrix, 2025-35 cycle, line for SOC 25-1022 (Mathematical science teachers, postsecondary): 57.1 thousand jobs in 2025, projected 58.1 thousand in 2035, a change of +1.0 thousand (+1.7%). Matrix employment counts all jobs in the occupation, including self-employed workers (0.1% of this occupation in 2025). The average for all occupations in this cycle is +3.5%.
Official statisticReport an error
How exposed is this work to AI (research measures)
Scores as published by each study. Each row says what it measures. These measure what AI could technically do, not forecasts of jobs lost.
Eloundou et al. (2023), "GPTs are GPTs"
2023 study · Share of tasks exposed to LLMs (β, human-rated)
43.2%
of tasks
What this measures

Published occupation-level exposure from the paper's occ_level.csv for O*NET-SOC 25-1022.00 Mathematical Science Teachers, Postsecondary. Shown: human-rated β = 0.432, the share of the occupation's tasks an LLM could do in at least half the time at equal quality, counting tasks that need LLM-powered software at half weight (β = E1 + 0.5 × E2). Other published measures for this occupation: human-rated α (LLM alone) 0.189 and γ (all LLM-assisted) 0.676; GPT-4-rated β 0.541. This measures how much of the work an LLM could touch, not how many jobs disappear.

Official statisticReport an error
Today, in this role

What's changing day to day for math professors

Three parts of your work where AI is already doing real lifting, and what stays yours.

AI is sitting alongside you hereGrade and provide feedback on undergraduate mathematics homework, problem sets, and exams (including calculus, linear algebra, differential equations, and intro statistics) using AI-assisted grading tools (Gradescope autograder for numerical/code assignments, AI-assisted answer grouping for written free-response) and CAS verification (Wolfram Alpha Pro checking exact symbolic answers) while maintaining expert judgment on proof-quality and mathematical reasoning.

Grade and provide feedback on undergraduate mathematics homework, problem sets, and exams (including calculus, linear algebra, differential equations, and intro statistics) using AI-assisted grading tools (Gradescope autograder for numerical/code assignments, AI-assisted answer grouping for written free-response) and CAS verification (Wolfram Alpha Pro checking exact symbolic answers) while maintaining expert judgment on proof-quality and mathematical reasoning.[11],[1]

Where your edge is

Adopt Gradescope for all problem sets that have numerical or symbolic answers: Gradescope's AI-assisted grouping clusters similar student solution approaches so you apply a rubric once per strategy family rather than once per submission. Reserve manual grading effort for the authentically hard judgment: is this proof argument valid? Is this student's reasoning correct even if the answer is wrong?

AI is sitting alongside you hereProvide adaptive instruction in introductory mathematics courses (calculus I-III, precalculus, and college algebra) using AI-powered adaptive learning platforms (ALEKS, Khanmigo) to handle personalized practice and immediate feedback outside class, freeing lecture time for conceptual development and problem-solving strategies that AI tutors cannot scaffold at the level of mathematical insight.

Provide adaptive instruction in introductory mathematics courses (calculus I-III, precalculus, and college algebra) using AI-powered adaptive learning platforms (ALEKS, Khanmigo) to handle personalized practice and immediate feedback outside class, freeing lecture time for conceptual development and problem-solving strategies that AI tutors cannot scaffold at the level of mathematical insight.[14],[15]

Tools picking this up
Where your edge is

Deploy ALEKS or Khanmigo for the adaptive practice component of introductory courses: ALEKS's knowledge-space engine correctly diagnoses which prerequisite gaps are blocking each student and generates targeted practice, reducing the number of "I don't understand this at all" office-hour cases that are really prerequisite deficits. Redirect class time to problem-solving approach modeling and the mathematically interesting "why does this method work?" questions that adaptive practice cannot answer.

AI is sitting alongside you hereHold office hours and respond to student questions on mathematical concepts, proof strategies, and problem-solving approaches, triaging which questions AI tutors (Khanmigo, Wolfram Alpha, ChatGPT) can resolve adequately, and reserving human engagement for the student who has tried AI-assisted problem-solving and still cannot resolve a deep conceptual block or proof-writing difficulty.

Hold office hours and respond to student questions on mathematical concepts, proof strategies, and problem-solving approaches, triaging which questions AI tutors (Khanmigo, Wolfram Alpha, ChatGPT) can resolve adequately, and reserving human engagement for the student who has tried AI-assisted problem-solving and still cannot resolve a deep conceptual block or proof-writing difficulty.[15],[3]

Where your edge is

Redirect office hours from "how do I compute this derivative?" (Wolfram Alpha answers that better than you can at 10pm) to "I understand the mechanics but don't see why the proof works": the genuinely hard questions that require mathematical intuition-building. Track which question types still escalate despite AI availability; those are your highest-value teaching moments and the data that should drive your lecture design.

Where math professors go next

Natural next steps for someone with your foundation: not exits, evolutions.

What each move takes

A direction you could grow

Education Administrators, Postsecondary

Mathematics faculty frequently move into department chair, associate dean of science, or dean of arts-and-sciences roles, particularly those who have led curriculum redesign efforts, chaired accreditation self-studies, or served on institution-wide AI governance committees. Mathematics departments are under more curricular pressure than almost any other discipline in 2025–2026 (calculus sequence reform, AI tool policy, proof course restructuring), creating demand for administrators who understand the substance of what needs to change rather than just the process. The pivot is natural for faculty who have demonstrated governance engagement and want institutional impact beyond their own research program.

What you'd add
  • · Higher education budget management: faculty line planning, equipment requests, and grant indirect-cost negotiation
  • · Accreditation processes: regional and disciplinary accreditation self-study coordination (HLC, SACSCOC)
  • · Faculty performance review, promotion/tenure facilitation, and hiring committee leadership
  • · Enrollment management basics: major declaration trends, STEM pipeline programs, and retention data interpretation
  • · Institutional AI governance: developing student AI-use policy, faculty development programs, and vendor evaluation frameworks for AI tools in STEM courses
What it takesSome new skills to pick up
Tools of the era

How has the work of math professors changed since 1727?

The tools that defined the work, one line per era. Select an era to see how it reshaped the work.

  • Blackboard, chalk, and published mathematical tables (colonial and antebellum era)

    For more than a century of the role, a mathematics professor's entire pedagogical toolkit was the blackboard, chalk, and printed mathematical tables. Blackboards became standard in US colleges in the early 1800s (West Point introduced them around 1817, and the practice spread rapidly to other institutions). Before that, mathematical instruction was conducted through oral recitation of assigned textbook passages: students memorized and recited, the professor corrected. The blackboard shifted the mode from recitation to demonstration: the professor could now work problems in front of the class and show each step, a change that restructured how mathematical ideas were communicated. Printed tables of logarithms, trigonometric values, and astronomical data were the computation aids; a faculty member who could use them rapidly was considered technically proficient.

  • Research seminar model + slide rule + mimeograph (German research university era)

    The founding of Johns Hopkins University in 1876 transplanted the German research seminar into American mathematics. Under J. J. Sylvester, graduate students met weekly to discuss original research problems alongside their coursework. This changed what mathematics faculty were expected to do: before Hopkins, a math professor taught undergraduates and perhaps wrote a textbook; after Hopkins, the expectation of original publication became a condition of employment at research universities. The slide rule (popularized from the 1880s onward) gave engineering and applied math students a portable analog computation tool; mimeograph and later ditto machines (from the 1920s) let faculty reproduce problem sets and course notes at scale for the first time. Neither tool transformed teaching as much as the seminar model did: the seminar shifted the professor from transmitter of known mathematics to active researcher whose research enriched the teaching.

    Effect on the work

    The research expectation created a bifurcated workforce: research university faculty (expected to publish) and teaching-only faculty at liberal arts colleges and normal schools (expected only to teach). This tension persists today in the form of the tenure-track vs. adjunct split.

  • Federal funding + "New Math" curriculum reform (GI Bill, Sputnik, SMSG era)

    The GI Bill (1944) and the Sputnik crisis (1957) were not merely demographic and political events: they were technology-and-curriculum events for mathematics faculty. The GI Bill doubled college enrollment and forced rapid faculty hiring; Sputnik triggered the National Defense Education Act (1958), which poured federal funding into university science and mathematics programs and created large summer institutes for faculty retraining. The School Mathematics Study Group (SMSG), founded in 1958 and funded by the NSF, assembled university mathematics professors to rewrite the K-12 and college mathematics curriculum around modern set theory and abstract algebra. Hundreds of college faculty participated in SMSG institutes; the "New Math" they produced restructured how calculus and introductory mathematics were taught at the university level for the next two decades. Faculty who trained under this model brought Bourbaki-style abstract rigor into courses that had previously been more computational.

    Effect on the work

    NSF summer institutes trained thousands of faculty in new mathematical approaches. The NDEA dramatically increased federal funding for mathematics departments and created new faculty positions.

  • Scientific calculator (TI-30, 1976; HP-15C, 1982) and programmable calculator era

    Texas Instruments released the TI-30 scientific calculator in 1976 at a consumer price low enough for college students to own personally. Within three years the slide rule had effectively vanished from college mathematics classrooms. The shift was not trivial: for nearly a century, a mathematics professor's pedagogical choices had been shaped by the slide rule's inherent imprecision (a few significant figures at best). The scientific calculator's ability to produce several digit results changed what numerical mathematics was taught and how problems were framed. Programmable calculators (HP-15C, 1982; TI-81, 1990) took this further, enabling students to code iterative algorithms in their palms. Faculty who had never had to decide whether to allow calculators on exams suddenly faced a structurally new question that had no obvious right answer, and that question has not been resolved forty years later.

  • Computer algebra systems: Wolfram Mathematica (1988), Maple (1985 commercial), MATLAB

    Wolfram Mathematica launched on June 23, 1988, and was immediately recognized as transformative: for the first time, a single piece of software running on a workstation could perform symbolic integration, differentiation, matrix decomposition, differential equation solving, and visualization with a unified interface. Stephen Wolfram described Mathematica 1.0 as achieving "a wow that one could routinely do integrals symbolically by computer." Maple had been available commercially since 1985 (Waterloo Maple); MATLAB was developed at Stanford from 1984 for numerical computation. For mathematics faculty, CAS software posed a genuine pedagogical challenge: if a computer can solve a first-year calculus problem instantly, what should a first-year calculus course be teaching? The debate between "CAS-active" and "CAS-neutral" instruction defined a generation of mathematics pedagogy discussions, and it remains unresolved. Faculty who adopted Mathematica for research found it transformed how they explored conjectures, generated examples, and verified results.

    Effect on the work

    CAS software began the slow erosion of the "computation" component of mathematics coursework. Departments that adopted Mathematica or Maple for instruction shifted course objectives from computational fluency toward conceptual understanding and proof, requiring faculty to rebuild their courses from scratch.

  • Online homework systems and adaptive learning: WebAssign, WeBWorK, ALEKS

    The 2000s and 2010s saw the widespread adoption of online homework and adaptive learning platforms in collegiate mathematics. WeBWorK, an open-source system developed at the University of Rochester from the mid-1990s, was deployed at hundreds of institutions and automated the grading of numerical and symbolic homework problems. ALEKS (Assessment and Learning in Knowledge Spaces), founded in 1994 and adopted widely after McGraw-Hill acquired it in 2013, used knowledge-space theory to adaptively diagnose each student's mathematical gaps and generate personalized practice queues. For faculty, these tools shifted the grading burden dramatically: a calculus section of hundreds of students could have every problem set auto-graded, freeing instructor time from routine marking toward conceptual engagement. The downside was that online homework optimized for answer-getting rather than mathematical reasoning, a tension faculty navigated through the decade.

    Effect on the work

    Online homework systems reduced the grading workload for large introductory courses, but also enabled departments to increase class sizes without proportional faculty additions. The adjunct share of mathematics teaching continued to rise through this era, partly because the routine grading that justified TA allocation was now automated.

  • AI mathematics tools: GPT-4 + Wolfram Alpha Pro, AlphaGeometry, DeepSeek-Math, OpenAI o3, ALEKS + Khanmigo

    The 2022-2026 period brought a qualitative shift in AI capability for mathematics that directly challenged the content of postsecondary mathematics courses. OpenAI o3 scored 96.7% on AIME 2024 competition problems (December 2024), problems designed to be among the hardest in high school mathematics and previously unsolvable by AI systems. Google DeepMind's AlphaGeometry (Nature, January 2024) solved International Mathematical Olympiad geometry problems at silver-medal level. DeepSeek-Math (arXiv 2402.03300, 2024) demonstrated that large language models fine-tuned on mathematical data could produce competition-quality solutions. Wolfram Alpha Pro, combined with ChatGPT, now solves the vast majority of undergraduate mathematics homework problems on demand. For mathematics faculty, this is the most structurally disruptive technology shift in the occupation's history: not because AI can replace the professor, but because it invalidates most of the traditional course assessment formats. A calculus exam question that asks "evaluate the integral" is answered correctly by Wolfram Alpha in three steps. The MAA (Mathematical Association of America) has documented the shift explicitly: departments are now redesigning calculus sequences around proof construction, error identification in AI outputs, and oral mathematical defenses rather than routine computation.

    Effect on the work

    The Federal Reserve FEDS Notes (February 2026) found that Mathematics and Computer Science majors have the highest AI language-model exposure score among all college major categories, suggesting that the demand for mathematics instruction itself may be affected as AI reduces the skill premium for mathematical computation fluency.

Part of Education · see all 51 roles →
Different role?

See the same long-arc view for your own profession.

Browse the directory by industry, or search by title or SOC code.

Browse all roles

Mathematical Science Teachers, Postsecondary employment and pay by year

On record since1727
Latest US employment47,670 (May 2025)Source: BLS OEWS May 2025Official statisticReport an error
Latest US median pay$79,940 (May 2025)Source: BLS OEWS May 2025Official statisticReport an error
Outlook+1.7% by 2035Source: BLS Employment Projections 2025-35Official statisticReport an error
View all 29 years of cited US data, each linked to its source
YearUS employmentAnnual pay (median unless marked)SourceReport
18701,200n/aEst., NCES Digest of Education Statistics Researched contextReport
19003,500n/aEst., AMS Researched contextReport
195012,000n/aEst., NCES Researched contextReport
1961n/a$10,000Est., AAUP Salary Report, January 1961 Researched contextReport
197028,000n/aEst., NCES Researched contextReport
199545,000$45,000Est., NCES (employment) Researched contextEst., NCES (pay) Researched contextReport
200341,880$50,910BLS OEWS May 2003 Official statisticReport
200443,760$52,450BLS OEWS May 2004 Official statisticReport
200544,660$53,820BLS OEWS May 2005 Official statisticReport
200644,570$56,420BLS OEWS May 2006 Official statisticReport
200744,560$58,560BLS OEWS May 2007 Official statisticReport
200845,710$61,120BLS OEWS May 2008 Official statisticReport
200948,100$63,640BLS OEWS May 2009 Official statisticReport
201051,980$65,710BLS OEWS May 2010 Official statisticReport
201153,650$66,680BLS OEWS May 2011 Official statisticReport
201253,530$64,990BLS OEWS May 2012 Official statisticReport
201353,970$65,930BLS OEWS May 2013 Official statisticReport
201454,010$65,190BLS OEWS May 2014 Official statisticReport
201553,850$67,170BLS OEWS May 2015 Official statisticReport
201652,020$69,520BLS OEWS May 2016 Official statisticReport
201750,870$70,910BLS OEWS May 2017 Official statisticReport
201851,250$73,230BLS OEWS May 2018 Official statisticReport
201951,150$73,690BLS OEWS May 2019 Official statisticReport
202049,550$73,650BLS OEWS May 2020 Official statisticReport
202144,140$77,580BLS OEWS May 2021 Official statisticReport
202246,560$77,420BLS OEWS May 2022 Official statisticReport
202348,230$81,020BLS OEWS May 2023 Official statisticReport
202448,820$79,350BLS OEWS May 2024 Official statisticReport
202547,670$79,940BLS OEWS May 2025 Official statisticReport
Embed this timeline on your site

Free for any site. Paste this where the timeline should appear; it stays interactive, every datapoint stays cited, and it sets no cookies on your page. How embedding works

<iframe src="https://futurehistory.earth/embed/25-1022"
  width="100%" height="430" style="border:0"
  title="Mathematical Science Teachers, Postsecondary, a Future History timeline"
  loading="lazy"></iframe>
<p style="font-size:13px;margin:4px 0 0"><a href="https://futurehistory.earth/timeline/25-1022">Mathematical Science Teachers, Postsecondary: jobs and pay over time</a>, from Future History</p>

See all roles in Education