Tutoring Center Video Statistics

Key takeaways

1. Of YouTube’s 100 most-viewed results for ‘tutoring’ on November 7, 2026, 82 were about tutoring, but only 1 of the 68 channels behind them sold tutoring and none was a tutoring center; comedy, dramas and TV scenes about tutors made 51.2% of the on-topic videos and drew 39.9% of their views (our read). (Mark Studios) See details
2. Seven tutoring center franchisors and chains posted 182 videos to their official YouTube channels in the year to November 7, 2026, with a median of 224.5 views; across 14 tutoring and test-prep brand channels the median upload had 893 views (our read). (Mark Studios) See details
3. Across 96 randomized trials, tutoring raised preK-12 learning by a pooled 0.37 standard deviations (NBER, 2020), revised to 0.288 in the 2024 journal version; in a 2024 meta-analysis, tutoring delivered virtually pooled to 0.07 standard deviations against 0.44 for in-person programs. (NBER) See details
4. Parents said 24% of U.S. K-12 children got academic tutoring in 2024-25 (50CAN, fall 2025); 78% of public schools had provided some tutoring by October 2024, and principals estimated 11% of students got high-dosage tutoring in 2023-24 (NCES). (NCES) See details
5. BLS counted 175,070 wage-and-salary tutor jobs in May 2025 at a median wage of $43,350, 15.0% below the median for all occupations, and 208.4 thousand tutor jobs in 2025 including the self-employed. (BLS) See details
6. Private exam preparation and tutoring establishments averaged 11,393 in 2025, up 17.1% from 2019 (BLS QCEW), and employer firms took in $7.91 billion in 2024 ($7,907,993,000), according to the Census Bureau. (BLS) See details
7. Nine tutoring franchise systems reported 4,029 U.S. centers, 297 of them in Florida; the seven with 2026 disclosure documents went from 3,908 centers at the start of 2023 to 3,919 at the end of 2025 (our count). (Mark Studios) See details
8. The much-repeated “video lifts landing-page conversions 80%” traces to EyeView, a video vendor, in 2009, and one of the tests behind it was a single landing page at TutorVista, an online tutoring company: EyeView’s case study said 86 percent and its blog quoted “over 80 percent”. (EyeView) See details

What does YouTube show people who search for 'tutoring'?

Mostly entertainment about tutors, not tutoring businesses. In our November 7, 2026 read of YouTube’s most-viewed results for 'tutoring', comedy, sketches, dramas, TV and film scenes and role-plays about tutors made 51.2% of the on-topic videos and drew 39.9% of their lifetime views, while lessons that teach a subject made 39.0% of the videos and drew 23.4% of the views (our count). Only 1 of the 68 channels behind the results belonged to a company that sells tutoring, the online marketplace Wyzant; no tutoring center, franchise or other local tutoring business appeared. We read YouTube’s own view-sorted results for 'tutoring' through the YouTube Data API with no country filter, the first 100 results, and classified every channel by hand from its own title, description and video text. YouTube’s daily search limit for our API key stopped the read there, before the searches we had planned for 'tutor' and 'SAT prep', so this is one search on one day, not a census.

The results had 920,248,560 lifetime views in all. 18 of the videos used 'tutor' in another sense, and they drew 30.1% of the views: 5 were versions of one Punjabi song titled 'Tutor', and 11 were Indonesian makeup, cooking, drawing and other how-to videos, because 'tutor' is Indonesian shorthand for tutorial. YouTube ranked the official video of that Punjabi song first, with 113,225,023 views, although an AI-tutor ad lower in the list had more. Leaving those videos out leaves 82 on-topic videos with 643,090,495 views. View counts are lifetime snapshots and include views from ads.

Who makes the most-viewed tutoring videos on YouTube?

Tutors and education brands make under half of them. Tutoring sellers, edtech brands, individual tutors and teachers, and one education nonprofit made 46.3% of the on-topic videos among YouTube’s most-viewed results for 'tutoring' on November 7, 2026, and drew 58.5% of the views (our count). Individual tutors and teachers posting as themselves made 35.4% of the videos and drew 20.3% of the views. Teacher Aya Online Tutor, a Philippine channel of Tagalog and English reading and math lessons for young children, posted 18 of the videos, more than any other channel. The Organic Chemistry Tutor, a U.S. channel of math, chemistry and physics lessons, posted 7, the second most. We did not check any tutor’s or teacher’s credentials.

Entertainment channels make the rest. Media, TV, film, podcast and short-drama studio channels made 19.5% of the on-topic videos and drew 17.6% of the views, and comedians, skit makers, animators and other entertainment creators made 34.1% of the videos and drew 23.9% of the views (our count). Their videos include Young Sheldon and Modern Family scenes, Saturday Night Live’s 'Tutoring the Cool Kid', stand-up sets, 'me as a private tutor' skits and short dramas about a new tutor.

Which tutoring videos get the most views?

An AI-tutor ad. The most-viewed video among YouTube’s most-viewed results for 'tutoring' on November 7, 2026 was the nonprofit CK-12 Foundation’s short ad for its Flexi AI tutor, with 214,960,145 views, and that one ad drew 33.4% of all on-topic views (our count). Wyzant’s ad 'Online Tutoring That Fits Your Family' had 3,917,843 views. Ads, product demos and app lists for a tutoring service, an AI tutor or a teaching tool made 4 of the on-topic videos, and 4 more were about tutoring as work: how much online tutors earn, where to find online tutoring jobs, and what South Korea’s star cram-school tutors make.

A few channels hold most of the views. The three channels with the most views, the CK-12 Foundation, Teacher Aya Online Tutor and the prank channel Cameron Whitt, drew 58.1% of the on-topic views, and the ten biggest drew 80.0% (our count).

How long, how new and in what language are the most-viewed tutoring videos?

Most run longer than a minute. Videos of 60 seconds or less were 31.7% of the on-topic videos among YouTube’s most-viewed results for 'tutoring' on November 7, 2026 and drew 33.3% of the views, and 23.2% ran longer than 20 minutes, most of them lessons (our count). 59.8% of the on-topic videos were published in 2024 or later.

English-language videos made 73.2% of the on-topic videos, and 11 were in Filipino, nearly all of them Teacher Aya Online Tutor’s Tagalog reading lessons. Channels whose country setting was the United States drew 54.8% of the on-topic views; 20 of the channels had their country set to the United States, and none named a place in Florida in its own text.

25 of the videos were marked as made for kids, all of them lessons from Teacher Aya Online Tutor, Tutoring Hour and The Organic Chemistry Tutor. YouTube’s altered or synthetic content field was absent from every one of the 100 videos, the CK-12 AI-tutor ad included.

How often do tutoring and test-prep brands post to YouTube?

Rarely, for most of them. The official YouTube channels of the tutoring and test-prep brands we read posted 647 videos in all in the year to November 7, 2026, and the median channel posted 9 (our count, every upload in the year read through the YouTube Data API). We found official channels for 15 of the brands we looked up, counting Khan Academy and the College Board as nonprofit comparators; Club Z! Tutoring, Best in Class Education, Eye Level and The Tutoring Center had no channel we could confirm, and Tutor Doctor’s only channel we found is for selling franchises. 3 of the tutoring and test-prep channels posted nothing in the year: Tutor Doctor’s franchise-sales channel had not posted since 2015, C2 Education’s channel since 2013 and Kaplan Test Prep’s main channel since 2021.

Their videos reach few people. The tutoring and test-prep channels’ videos from the year had a median of 893 views, and 57.2% ran 60 seconds or less. The seven tutoring center franchisors and chains, Kumon North America, Mathnasium, Sylvan Learning, Huntington Learning Center, JEI Learning Centers, Tutor Doctor and C2 Education, posted 182 videos in the year, with a median of 224.5 views. Only 1 of the channels in the table, Wyzant’s, had a video among YouTube’s most-viewed results for 'tutoring'.

The tutoring and test-prep channels had 786,497 subscribers between them. Khan Academy, a free nonprofit, had 9.51 million (9,510,000) on its own, and 18 of its uploads in the year named AI or its Khanmigo AI tutor in the title; none of the tutoring and test-prep channels’ uploads did.

YouTube rounds subscriber counts, and view counts include views from ads. Each brand’s regional, franchise-location and subject channels are not counted. The per-channel data, including total videos, lifetime views and the share of each channel’s videos that ran 60 seconds or less, are in our brand channel table (CSV); the search sample is in youtube-videos.csv and youtube-channels.csv.

How much does tutoring raise test scores?

Tutoring raised preK-12 learning outcomes by a pooled 0.37 SD across 96 randomized trials in Nickow, Oreopoulos and Quan’s 2020 meta-analysis, NBER Working Paper 27476 (NBER). The peer-reviewed version of the same review, published in the American Educational Research Journal in 2024, reports a smaller pooled effect of 0.288 SD (AERJ abstract via ERIC), 22% below the working-paper figure (our calculation). Most of the trials were in elementary grades.

In percentile terms, the average tutored student scores above about 64% of comparison students on the 2020 figure and above about 61% on the 2024 figure, assuming normally distributed scores (our calculation). J-PAL’s 2020 evidence review translated the 2020 figure as moving a median student to nearly the 66th percentile (J-PAL).

Effects shrink as programs grow. Kraft, Schueler and Falken’s 2024 meta-analysis, Annenberg EdWorkingPaper 24-1031, covered 282 randomized trials published from 1967 to 2023 (EdWorkingPapers) and found a pooled 0.42 SD across its full sample. For U.S. programs serving a thousand students or more, measured on standardized tests, its preferred estimate is 0.16 SD (full paper), 38% of the full-sample figure (our calculation). That would put the average tutored student above about 56% of comparison students (our calculation). The authors note that few trials of programs that large exist, so the estimate is imprecise.

How do the tutoring trials compare side by side?

The trials on this page range from a 0.37 SD drop in final grades with an AI tutor in a 2026 university trial to an estimated 0.73 to 1.3 SD gain with a custom AI physics tutor in 2025; large school-based trials of in-person human tutoring fall between 0.16 SD and 0.37 SD. Effects are in standard deviations (SD) of test scores unless stated. Each figure links to its paragraph below.

Which tutors, grades and schedules get the biggest effects?

Programs run by certified teachers did best in Nickow, Oreopoulos and Quan’s 2020 meta-analysis: a pooled 0.50 SD, against 0.40 SD for paraprofessionals such as school staff and education students, 0.21 SD for volunteers and 0.23 SD for parents (NBER paper, Table 3A). The teacher figure is 2.4 times the volunteer figure (our calculation). The authors note that the teacher advantage rests largely on first-grade Reading Recovery studies.

Effects were smallest in the older grades: tutoring in grades 6-11 had a pooled effect of 0.16 SD, from a small number of trials (NBER paper, Table 3B). More sessions a week went with larger effects: 0.24 SD for programs meeting one or two days a week and 0.41 SD for four or five days, 1.7 times as large (our calculation). The 2024 journal version names the same pattern: effects were largest for programs that use teachers or paraprofessionals, serve earlier grades, meet at least three days a week and run during the school day.

What did the largest U.S. tutoring trials find?

Daily in-school math tutoring by Saga Education raised Chicago students’ math test scores by 0.16 SD in a first randomized trial of 2,633 ninth and tenth graders, and by 0.37 SD in a second trial of 2,710 students (Guryan et al., NBER Working Paper 28531, 2021) (NBER). The authors put the model’s cost at $3,500 to $4,300 per participant per year (NBER).

A cheaper design held up in a trial of about 4,000 students in two large districts in 2018-19 and 2019-20: groups of four that split time between an in-person tutor and computer-assisted practice gained 0.23 SD in math (Bhatt et al., NBER Working Paper 32510, 2024) (NBER). The authors call that almost as large as the two-student model and cheaper per pupil.

Scaling tutoring nationally would be expensive. Kraft and Falken’s 2021 blueprint, Annenberg EdWorkingPaper 20-335, estimated that targeted school-wide tutoring, such as in K-8 Title I schools, would cost between $5 and $15 billion a year (EdWorkingPapers).

Does online or video-call tutoring work as well as in-person tutoring?

Online tutoring trials have so far produced smaller effects on average. In Kraft, Schueler and Falken’s 2024 meta-analysis, the virtual-tutoring estimates pooled to 0.07 SD, against an unadjusted 0.44 SD for in-person programs (EdWorkingPaper 24-1031, full paper), about 16% of the in-person figure (our calculation). At 0.07 SD the average student tutored online would score above about 53% of comparison students (our calculation). The authors find that other program features likely drive most of that gap: holding those features constant, the difference between virtual and in-person programs was not statistically significant. Most of the virtual trials ran after the pandemic began.

Single trials vary widely. Carlana and La Ferrara’s 2020 trial in Italy randomized 1,059 middle school students who applied during the lockdown (J-PAL) and found that free one-on-one online tutoring by university volunteers raised academic performance by 0.26 SD (EdWorkingPaper 21-350, 2021) (EdWorkingPapers). In Spain, eight weeks of online math tutoring by qualified teachers for pairs of students raised test scores by 0.26 SD in a 2021 trial of 356 students (Gortazar, Hupkau and Roldán, EdWorkingPaper 23-743) (EdWorkingPapers). When a regional education authority ran the same program for 1,344 students, the effect on test scores fell to 0.11 SD, about one-third of the pilot’s effects by the authors’ account (Calle et al., EdWorkingPaper 26-1496, 2026) (EdWorkingPapers).

Lower-dosage U.S. programs found small effects. Online volunteer tutoring twice a week for 560 Illinois middle school students produced 0.07 SD in math, not statistically significant (Kraft, List, Livingston and Sadoff, EdWorkingPaper 22-568, 2022) (EdWorkingPapers). Virtual reading tutoring for 2,085 kindergarten to second-grade students raised early literacy skills by 0.05-0.08 SD, with one-on-one sessions doing better than pairs (Robinson, Pollard, Novicoff, White and Loeb, EdWorkingPaper 24-955, 2024) (EdWorkingPapers).

The one direct comparison found no difference. In a summer 2023 New York City program in which the same university-student tutors worked in person and remotely, literacy outcomes for the 446 students with complete test data did not differ significantly by format, though in-person students attended more (Hashim, Pace Miles and Croke, EdWorkingPaper 25-1176, 2025) (EdWorkingPapers).

Providers have moved online anyway. Of the 440 organizations listed in the National Student Support Accelerator’s High-Impact Tutoring Database on November 7, 2026, 310 offered virtual tutoring with a human tutor, such as over Zoom (NSSA), or 70% (our calculation). Listings are submitted by the organizations themselves, and one organization can offer several formats.

What have AI tutoring trials found?

Results depend on who uses the AI. In the 2024 Tutor CoPilot trial with 1,800 K-12 students, an AI assistant that gave human tutors expert-like guidance during live sessions left students 4 percentage points more likely to master topics (Wang, Ribeiro, Robinson, Loeb and Demszky, Stanford, arXiv preprint) (arXiv).

AI tutors that students use on their own have mostly shown small or negative effects in large trials. A two-year trial of Khan Academy’s Khanmigo in 18 Tennessee middle schools raised math achievement by about 0.06 to 0.08 SD over a school year, about the same as Khan Academy practice without AI; the authors report that students rarely engaged the tutor in real mathematical dialogue (Oreopoulos and Low, EdWorkingPaper 26-1551, 2026) (EdWorkingPapers). In a 2023-24 field experiment with nearly 1,000 high school math students in Turkey, students given unrestricted GPT-4 help during practice later scored 17% worse on an unassisted exam than students without it; a version with teacher-designed guardrails largely removed the harm (Bastani et al., PNAS, 2025) (PNAS via PMC). Among 2,379 undergraduates at a U.S. public university, access to a course-integrated AI tutor reduced final grades by 0.37 SD (Liu et al., EdWorkingPaper 26-1598, 2026) (EdWorkingPapers).

The largest positive result comes from one carefully designed tutor tested over two lessons. In a crossover trial in a Harvard physics course with 194 students, a custom AI tutor beat an in-class active-learning lesson by an estimated 0.73 to 1.3 standard deviations on a short post-test (Kestin et al., Scientific Reports, 2025) (Scientific Reports via PMC).

Does video help students learn?

Yes, modestly when it replaces other teaching and strongly when it is added to it, in a 2021 systematic review of 105 randomized trials in higher education by Noetel and colleagues in the Review of Educational Research (Noetel et al., 2021). Swapping video for existing teaching methods produced a small improvement, an effect size of g = 0.28. Adding video to existing teaching produced a large one, g = 0.80. In plain terms, the average student taught with video in place of other methods scored above about 61% of the comparison group, and the average student given video on top of other teaching above about 79%, by our conversion. The trials studied college students, not tutoring clients; our training video statistics and higher education video statistics pages cover video in those settings.

What does research say about private tutoring centers?

The number of U.S. private tutoring centers more than tripled between 1997 and 2022, from about 3,000 to 10,000, in Kim, Goodman and West’s count of center locations in Infogroup business listings (EdWorkingPaper 21-367, 2024 version) (EdWorkingPapers). More than half were in the top fifth of areas by income. These are business-listing counts, not the Census establishment counts in the industry section.

A 2026 study of what centers do for students found gains on college admission tests but not on state tests. Using Texas records on over five million public school students and the opening and closing of centers near their high schools, Chen estimated that each year of tutoring center attendance raised SAT scores by roughly 8 to 13 points, with no effect on state standardized tests or graduation, and with the gains concentrated among students who were not economically disadvantaged (EdWorkingPaper 26-1619, 2026) (EdWorkingPapers).

How many U.S. students get tutoring?

About one in four. Parents said 24% of U.S. K-12 children had received academic tutoring in the 2024-25 school year, up from 19% in 2023-24, in 50CAN’s second Education Opportunity Survey (50CAN). Edge Research ran it online for 50CAN, an education advocacy nonprofit that does not sell tutoring, from September 30 to December 1, 2025, with 23,104 parents and guardians in all 50 states and D.C., weighted to Census and NCES profiles of parents. The gap between low-income and high-income children narrowed to 8 points, from 12 points a year earlier (50CAN).

Smaller, monthly polls put the share lower. In EdChoice’s October 10-12, 2025 poll of 1,334 school parents, run by Morning Consult, 19% said their child was getting tutoring outside regular school hours that school year and another 24% were looking or planning to look soon (EdChoice). EdChoice, a school-choice advocacy group, estimates from its monthly series since 2021 that 15–20% of students get any tutoring in a typical month, with a pandemic peak of 24% in April 2021 (EdChoice).

Very few get tutoring of the kind the research tests. In USC’s Understanding America Study, a probability-based household panel, about 15% of more than 1,600 households said in December 2022 that their student got any tutoring in the fall 2022 semester (USC). The share was 24% among students whose parents said they usually earned C’s or lower (USC). Only 11% of the tutored students had programs that met a moderate high-quality test (in person, four or fewer students per tutor, 30 minutes or more, three or more times a week) (USC), so across all the households, less than two percent of students got that kind of tutoring (USC). The schools section of this page has NCES’s School Pulse Panel counts, which measure what schools offer rather than what parents report.

Florida ran above the national rate in the first 50CAN survey. Parents said 23% of Florida K-12 children had received academic tutoring in the 2023-24 school year, against 19% nationally, in 50CAN’s July 8 to August 22, 2024 survey of 400 Florida parents (50CAN Florida report). The state sample is small, and the second edition prints Florida’s 2025 rate only as a chart bar, so we do not quote it.

The figures differ because the questions differ: a whole past school year against this month or this semester, any tutoring against tutoring outside school hours, children against households, and opt-in panels against probability panels. None of them separates paid private tutoring from free school programs.

Which families want tutoring but do not get it?

Nearly half of families without it. In 50CAN’s fall 2025 survey, 47% of families with no child being tutored said they were interested in tutoring (50CAN). Low-income families without tutoring who want a tutor named cost (31%) and transportation (16%) as the two largest barriers (50CAN). Free tutoring for students who fall behind was the most popular of the seven policies 50CAN tested: 86% of parents favored it strongly or somewhat (50CAN).

Interest splits by school type and place. In EdChoice’s combined May to July 2025 polls, interest in tutoring (already tutored or looking for a tutor) exceeded 55% among private-school parents and parents of children in special education, but was under one-third among rural and small-town parents (EdChoice).

Many parents do not think their child needs help. In Gallup’s 2023 survey for Learning Heroes of nearly 2,000 K-12 parents, 88% believed their child was at or above grade level in reading and 89% in math, while national test data put fewer than half of children at grade level (Gallup and Learning Heroes). The report, by the nonprofit Learning Heroes, is titled “B-flation: How Good Grades Can Sideline Parents”.

What do families pay for tutoring?

School parents said they would be able and willing to spend an average of $357 a month per child on tutoring outside school hours in EdChoice’s July 10-13, 2025 poll of 2,251 adults (EdChoice). That is stated willingness to pay, not spending.

Sellers publish starting prices, not averages. Sylvan Learning, a tutoring franchisor whose centers are locally owned and set their own prices, listed typical rates in November 2026 of $50-plus a session for personalized tutoring and homework help (Sylvan Learning) and $900-plus a class for ACT and SAT group classes (Sylvan Learning). Preply, an online marketplace where tutors in many countries set their own rates, said in November 2026 that the average price of its math lessons was $17 (Preply). A Sylvan starting rate and a Preply average are different statistics for different lesson lengths and markets, so we do not compare them. Franchise investment costs and fees are in the filings section of this page.

How many students take the SAT and ACT?

About 2.0 million (2,004,965) students in the U.S. high school class of 2025 took the SAT, College Board’s own count (College Board total group report), up from 1.97 million in the class of 2024 and the first class above 2 million since 2020 (College Board). Most took it in school: 1.4 million (1,361,927), 68% of takers, sat the SAT on a school day through a state, district or school program (College Board). The other 643,038, 32%, tested only on national weekend dates, the group that chose to register (College Board); they averaged 1173 against 961 for school-day takers, but weekend takers chose to register themselves, so the gap is not a measure of prep.

ACT counted 1.4 million (1,380,130) ACT-tested graduates in the class of 2025 (ACT profile report), up from 1.3 million (1,295,349) in the class of 2021 (ACT), a 6.5% rise (our calculation). Many students take both tests, so the two counts overlap. ACT says students who took the ACT more than once raised their ACT Superscore by an average of 2.4 points; the Superscore combines a student’s best section scores across sittings (ACT).

Florida had 204,840 SAT takers in the class of 2025, 87% of its 234,996 high school graduates (College Board Florida report), or 10.2% of all U.S. SAT takers (our calculation). Most Florida takers tested in school; only 27,614 tested on weekend dates alone (College Board), and Florida’s takers averaged 970 (College Board), against a national mean of 1029 (College Board total group report). Counting the PSAT tests as well, 694,009 Florida students took an SAT Suite test in 2024-25 (College Board). ACT estimates that 39% of Florida’s 2025 graduates took the ACT, against 36% nationally (ACT state table).

Do colleges and families still care about SAT and ACT scores?

Most colleges in Kaplan’s 2026 survey of admissions officers, 88% of them, will not require the SAT or ACT for 2026-27 (Kaplan), but most officers at those colleges say a strong score helps. Kaplan surveyed 245 admissions officers from July 14 to September 8, 2026, and 64% of officers at test-optional colleges (135 of 211) said a competitive score helps an applicant’s chances (Kaplan). Kaplan sells SAT and ACT prep.

Families who answer test-prep sellers’ surveys mostly test. In The Princeton Review’s 2026 College Hopes & Worries Survey, 93% of 9,446 college applicants and parents said they (or their child) took the SAT, the ACT or both (The Princeton Review). The Princeton Review sells prep and tutoring, and it promotes the survey to users of its own resources with a $5,000 prize drawing, so the sample leans toward families who prep. In a YouGov survey of 843 parents of 10- to 18-year-olds for Huntington Learning Center, fielded March 3-12, 2025, 79% agreed that supplemental learning such as tutoring is essential for preparing for college admissions (Huntington Learning Center).

One seller’s advertised score gains are larger than ACT’s national retest figure. Huntington’s home page advertises average gains of 229-plus SAT points and 5.4-plus ACT points for its students, with no sample, period or comparison group (Huntington, archived page), against the 2.4-point average Superscore gain ACT reports for all retesters. The research section of this page covers what controlled studies find.

How big do tutoring companies say they are?

Kumon, a Japanese franchisor of after-school math and reading centers, reports 3.51 million subject enrollments worldwide as of June 2026, 1.29 million in Japan and 2.23 million overseas (Kumon Group). A child taking math and reading counts twice. It reports 23,200 Kumon Centers worldwide as of March 2026 (Kumon Group), down 6.6% from 24,850 in December 2014 (Kumon fact sheet; our calculation). Kumon’s December 2014 press fact sheet, the most recent U.S. count we found on Kumon’s own site, listed 1,485 U.S. centers and 265,899 U.S. students (Kumon fact sheet); the filings section of this page has the current U.S. and Florida center counts from Kumon’s franchise disclosure document.

Online sellers give cumulative totals. Varsity Tutors, owned by Nerdy, advertises more than 10 million tutoring hours delivered and lists 1,000+ schools and universities, undated and without a method (Varsity Tutors). Nerdy’s audited figures are in the filings section. Wyzant’s rate and tutor-count pages were not used because wyzant.com’s robots.txt blocks the GPTBot AI crawler.

How do students use video and AI for help with schoolwork?

YouTube counts learning views in the billions. It says there were more than 5.5 billion views of learning and how-to content in the United States in June 2025 (YouTube). In December 2025, more than 20 million people used its AI “Ask” button on academic videos to simplify terms or summarize takeaways, a feature YouTube files under the heading “Your personal tutor” (YouTube). U.S. students watched more than 40 million hours through the ad-free YouTube Player for Education inside tools such as Google Classroom, Canvas and EdPuzzle in the first half of 2025 (YouTube), and YouTube cites a 2025 Oxford Economics study, without its sample, in which 79% of U.S. teachers who use YouTube agreed it helps students learn (YouTube). These are YouTube’s own unaudited figures about its own platform. In Google’s older 2017 survey of 918 U.S. monthly YouTube users aged 18-54, 86% said they regularly turn to YouTube to learn something new (Think with Google) and more than 7 in 10 said they use it for help with a problem in their work, studies or hobbies (Think with Google).

Free official SAT prep reached more students than paid prep in 2019, by the program owners’ count. Khan Academy and College Board said in January 2020 that more than 10 million students had signed up for free Official SAT Practice, a video-and-practice program, in its first five years (Khan Academy). In a 2019 survey of more than 65,000 SAT takers that the two cite, nearly four times as many students said they practiced with it as paid for commercial test prep (Khan Academy).

Just over half of U.S. teens have used AI chatbots for schoolwork help. In Pew Research Center’s fall 2024 survey of 1,391 U.S. teens aged 13 to 17, 26% said they had used ChatGPT for schoolwork, double the 13% of 2023 (Pew Research Center). In Pew’s September 25 to October 9, 2025 survey of 1,458 teens, 54% had used AI chatbots to get help with schoolwork (Pew Research Center), and One-in-ten said they do all or most of their schoolwork with a chatbot’s help (Pew Research Center). In the Gallup-Walton Family Foundation Voices of Gen Z survey of February 24 to March 4, 2026, 56% of Gen Z K-12 students used generative AI at least weekly (Gallup), but only 28% said their school gives them AI tools for schoolwork (Gallup). Districts moved tutoring online too: in RAND’s spring 2024 American School District Panel survey of 190 districts, 68% of large districts (10,000 or more students) offered online tutoring in 2023-24, against 17% of small districts (under 3,000) (RAND). General figures on children’s viewing and on online courses are on our kids video statistics and online course video statistics pages.

How many U.S. public schools offer tutoring?

In October 2024, 78% of U.S. public schools had provided some type of tutoring in the 2024-25 school year, down from 82% in October 2023, according to school leaders' answers to the National Center for Education Statistics' School Pulse Panel, a monthly survey of a national sample of public K-12 schools (NCES). As of October 2024, 22% of public schools provided no tutoring at all (NCES). By June 2025, 85% of public schools said they had provided tutoring at some point in the 2024-25 school year (NCES), 7 percentage points more than in October (our calculation), because the June question counts any tutoring given during the year.

High-dosage tutoring is the intensive kind. NCES defined it for respondents as tutoring of the same students three or more times a week for at least 30 minutes a session, one-on-one or in small groups, by trained tutors, tied to the curriculum and not drop-in homework help (NCES). It has held at a little over a third of public schools: 37% in December 2022 (NCES), 39% in October 2023 (NCES) and 37% in October 2024, a change NCES did not find statistically significant (NCES). Among schools that tutored in October 2024, about 47.4% offered the high-dosage kind (our calculation).

The less intensive kinds shrank. Standard tutoring, with no minimum session length or frequency, fell from 59% of schools in December 2022 (NCES) to 52% in October 2023 (NCES) and 48% in October 2024 (NCES), a drop of 11 percentage points (our calculation). Self-paced tutoring, where students work on their own, usually online, went from 21% (NCES) to 14% (NCES) and then 15% (NCES). On-demand online tutoring, where a school contracts an outside online tutoring company that offers round-the-clock help, was first asked about in October 2023, when 8% of schools offered it (NCES); it was 8% again in October 2024 (NCES). Comparing start-of-year snapshots, the share of schools with any tutoring slipped 5 percentage points from December 2022 to October 2024 (our calculation).

Start-of-year and end-of-year answers measure different things. In 2023-24, 87% of schools had provided some tutoring by the end of the year against 82% offering it in October, 5 points more, and high-dosage tutoring went from 39% to 46%, 7 points more (our calculations). NCES calls the School Pulse Panel an experimental data product. As of an October 26, 2026 copy of its home page, the June 2025 collection was still the latest; NCES had posted no 2025-26 results (NCES).

How many public school students get tutoring at school, and does it help?

Principals estimated that about 11% of all U.S. public school students had received high-dosage tutoring from their school in the 2023-24 school year as of October 2023, according to NCES's School Pulse Panel; the same share had received standard tutoring, and about 1% had used an on-demand online tutoring program their school offered (NCES). More than a third of schools offered high-dosage tutoring, but it reached a much smaller share of students.

Among schools offering high-dosage tutoring in October 2023, 38% rated it “very” or “extremely” effective for improving student outcomes, a higher share than rated standard or self-paced tutoring that way (NCES). These are school leaders' own ratings, not measured results. A federal survey of teachers asked a different question about the 2020-21 school year: in a GAO survey of 2,862 public school teachers in June and July 2021, run by Gallup, 26% of teachers who used after-school or weekend tutoring said it helped at least half of their students recover lost learning (GAO).

The need has not gone away. School leaders estimated that 40% of students started the 2024-25 school year behind grade level in at least one subject (NCES), and that 31% were still behind at the end of that year (NCES). These are leaders' estimates, not test scores.

How much federal pandemic money could go to tutoring?

Congress appropriated nearly $190 billion for the Elementary and Secondary School Emergency Relief (ESSER) fund in three COVID-19 relief laws, the Government Accountability Office reported (GAO). The largest round, the American Rescue Plan's ESSER fund, provided nearly $122 billion (Federal Register), of which at least $109.7 billion had to go to school districts (GAO). Each district had to reserve at least 20% of its ARP ESSER money to address lost instructional time through evidence-based interventions (Federal Register), at least $21.9 billion nationally (our calculation). The law's own examples were summer learning, extended-day, afterschool and extended-year programs; tutoring was an allowed use, not a required one.

No national figure breaks out the ESSER dollars spent on tutoring. Through school year 2021-22, districts reported spending about $57.3 billion in ESSER funds (GAO), or 30.2% of the appropriation (our calculation). About 80% of that went to students' academic, social and emotional needs, the Education Department category that includes tutoring, and to keeping schools running (GAO). In GAO's district visits, officials from 15 of the 17 districts said they used ESSER money to add instructional time through afterschool programs, summer school or a longer school year (GAO).

The School Pulse Panel shows how much school tutoring leaned on that money. In October 2023, 70% of schools offering high-dosage tutoring paid for it at least partly with federal funds such as ESSER (NCES). The share of all public schools using federal funds for tutoring fell from 65% in 2023-24 to 46% in 2024-25 (NCES), a drop of 19 percentage points (our calculation), after the September 30, 2024 deadline to commit ARP ESSER funds. Among schools with no tutoring in October 2024, 48% cited a lack of, or cuts in, funding and 36% said they could not find staff to support tutoring (NCES).

What was the national push for tutors, and did it reach its goal?

The federal goal was 250,000 new tutors and mentors. The U.S. Department of Education set it when it launched the National Partnership for Student Success on July 5, 2022, with AmeriCorps and the Johns Hopkins Everyone Graduates Center (U.S. Department of Education). AmeriCorps said the partnership would help districts find ways to use American Rescue Plan money for it, including ARP ESSER funds and $20 million in Volunteer Generation Funds to help nonprofits recruit and manage volunteers for school roles (AmeriCorps).

In October 2024, AmeriCorps reported that an estimated 187,000 additional adults joined tutoring, mentoring and other support roles in public schools in 2022-23 and 136,000 more in 2023-24, 323,000 in all (AmeriCorps), or 129% of the goal (our calculation). The estimate comes from a RAND survey of principals and counts mentors, college and career advisers and other support staff as well as tutors. Schools still struggled to fill tutor jobs: in October 2023, 11% of public school tutor positions were vacant, among the highest vacancy rates for non-teaching jobs (NCES).

What drop in test scores was school tutoring meant to address?

National test scores fell during the pandemic, and NAEP, the National Assessment of Educational Progress run by NCES, measured it. On NAEP's long-term trend assessment, average math scores for 9-year-olds fell 7 points from 2020 to 2022, the first decline in the series (NAEP), and reading fell 5 points, the largest reading decline since 1990 (NAEP). For 13-year-olds, math fell 9 points between the 2019-20 and 2022-23 school years (NAEP). On the main NAEP in 2024, fourth-grade math was still 3 points below 2019 (NAEP) and fourth-grade reading was 5 points below 2019 (NAEP). These national averages are context for why schools bought tutoring; they do not measure what tutoring did.

How many tutors work in the United States?

Tutors held 175,070 wage and salary jobs in the United States in May 2025, according to the Bureau of Labor Statistics’ Occupational Employment and Wage Statistics (OEWS) survey of employers (BLS). The count covers SOC 25-3041 Tutors on a payroll in any industry. It leaves out self-employed tutors, including tutors paid as independent contractors through online marketplaces.

BLS Employment Projections, which add the self-employed, put the occupation at 208.4 thousand jobs in 2025 (BLS). Self-employed workers held 12.3% of those jobs (BLS), about 25.6 thousand, computed from BLS’s rounded figures (our calculation). The projections count about 33,300 more tutor jobs than OEWS does, mostly the self-employed (our calculation).

Tutoring centers are only one employer. OEWS industry estimates put 69,400 tutors in other schools and instruction (NAICS 611600), the industry group that holds tutoring and test-prep businesses along with arts, sports and language schools (BLS). That is 39.6% of tutors on a payroll (our calculation). Elementary and secondary schools, public and private, employed 36,560 tutors (BLS), or 20.9% (our calculation). BLS publishes no tutor estimate for exam preparation and tutoring businesses (NAICS 611691) alone.

How much do tutors earn?

The median annual wage for U.S. tutors was $43,350 in May 2025, per BLS OEWS (BLS). The median hourly wage was $20.84 (BLS) and the mean annual wage $48,050 (BLS). OEWS turns hourly rates into annual figures by assuming a full-time, year-round schedule, so a part-time tutor paid the median rate earns less than the annual median in a year.

The 90th percentile wage was 2.6 times the 10th percentile in May 2025 (our calculation). The median for all U.S. occupations was $50,980 (BLS), so the tutor median was 15.0% lower (our calculation). These figures cover tutors on a payroll only, not the earnings of self-employed tutors or the owners of tutoring businesses.

Will there be more tutor jobs by 2035?

BLS projects 208.0 thousand U.S. tutor jobs in 2035, against 208.4 thousand in 2025, in its 2025-35 Employment Projections (BLS). That is a change of -0.2%, a slight decline (BLS), while total U.S. employment is projected to grow 3.5% (BLS).

BLS still expects about 34.2 thousand openings for tutors each year, on average, over the decade (BLS). Almost all of them come from tutors who move to other jobs or stop working, not from growth. Yearly openings equal 16.4% of the occupation’s 2025 employment (our calculation).

How many tutors work in Florida and South Florida, and what do they earn?

Employers in Florida counted 12,220 wage and salary tutors in May 2025, per BLS OEWS (BLS), 7.0% of the U.S. total (our calculation). The Florida median annual wage for tutors was $37,310 (BLS), or $17.94 an hour (BLS). That is $6,040 below the national tutor median (our calculation). Florida’s median for all occupations was $47,880 (BLS), so tutors in the state earned 22.1% less at the median (our calculation). The top of the range is higher in Florida: the state’s 90th percentile tutor wage was $83,140 (BLS), above the national $75,990.

The Miami-Fort Lauderdale-West Palm Beach metro area, which covers Miami-Dade, Broward and Palm Beach counties, had 4,940 tutors on a payroll (BLS). That is 40.4% of Florida’s tutors (our calculation). The metro’s location quotient for tutors was 1.56 (BLS): tutors’ share of local jobs was 1.56 times their share of jobs nationally. The metro median tutor wage was $38,500 (BLS). The metro median for all occupations was $48,640 (BLS), so tutors earned 20.8% less at the median (our calculation).

OEWS publishes no Martin County figure on its own. Martin is part of the Port St. Lucie metro area with St. Lucie County, where employers counted 190 tutors (BLS) with a median annual wage of $36,190 (BLS).

How many tutoring and test-prep businesses are there in the U.S.?

The Bureau of Labor Statistics counted an annual average of 11,393 private exam preparation and tutoring establishments (NAICS 611691) in the United States in 2025, in its Quarterly Census of Employment and Wages (QCEW) (BLS). In 2019 the count was 9,730 (BLS), so the number grew 17.1% in six years (our calculation). QCEW counts worksites with employees covered by unemployment insurance: a chain counts once per location, and tutors who work alone with no payroll are not in it.

Those establishments employed 99,216 workers on average in 2025 (BLS), counting everyone on the payroll, not only tutors. Average annual pay was $31,500 (BLS). That average spreads total wages over every worker, part-time and full-time, so it sits well below a full-time salary.

The Census Bureau counts fewer establishments. County Business Patterns found 9,820 exam preparation and tutoring establishments with paid employees in 2023 (Census), with 108,755 employees in the March pay period (Census). That is an average of 11.1 employees per establishment (our calculation). The 2022 Economic Census counted 9,750 (Census). Census counts establishments from its business register, while QCEW averages state unemployment-insurance records over the year, so the two never match exactly.

How much revenue do U.S. tutoring businesses bring in?

U.S. exam preparation and tutoring establishments with paid employees had revenue of $7.29 billion in 2022, according to the 2022 Economic Census (Census). That is about $747,300 per establishment (our calculation). The Census Bureau’s Annual Integrated Economic Survey put revenue of tutoring and exam-prep employer firms at $7.91 billion in 2024 (Census), up 5.4% from 2023 (Census). The survey is a sample of firms, so its total is not directly comparable with the Economic Census count.

Tutoring and exam prep is one part of a bigger industry group. All U.S. employer firms in other schools and instruction (NAICS 6116), which also holds fine arts, sports, language and driving schools, took in $36.81 billion in 2024 (Census). Tutoring and exam-prep firms had 21.5% of that (our calculation). All these revenue figures leave out tutors with no employees.

Most of the money comes from tutoring itself. Academic tutoring and customized learning programs made up 57.1% of the industry’s 2022 revenue, per the Economic Census product lines (Census), and 80.6% of establishments reported revenue from them (Census). Exam preparation courses made up 27.3% of revenue (Census). The product lines describe what was sold, not how it was delivered: Census does not split online from in-person tutoring.

Not every tutoring business is a company run for profit. 1,397 of the establishments in the 2022 Economic Census were exempt from federal income tax (Census), 14.3% of the total (our calculation). Tax-exempt establishments are mostly nonprofits.

How many tutors and teachers run a business with no employees?

Census Nonemployer Statistics counted 924,679 educational services businesses with no paid employees in the United States in 2024 (Census), up from 893,520 in 2023 (Census). Most are sole proprietors filing business income with the IRS. Census publishes this count only for all of educational services (NAICS 611), so independent tutors are mixed in with solo music teachers, driving instructors, language teachers and other instructors.

In 2023 there were 7.6 of these education businesses without employees for every private education establishment with employees (our calculation), using County Business Patterns’ count of 118,004 educational services establishments that year (Census). Nonemployer education businesses had receipts of $17.33 billion in 2024 (Census), or about $18,746 each on average (our calculation). A nonemployer business can be a side business of someone who also holds a wage job. BLS’s separate estimate of about 25.6 thousand self-employed tutor jobs in 2025 is the closest figure for tutors alone.

How much do education businesses spend on advertising?

U.S. educational services employer firms spent $3.27 billion on purchased advertising and promotional services in 2024, per the Census Bureau’s Annual Integrated Economic Survey (Census). That covers every private school, college, training firm and tutoring business with employees (NAICS 61), and it counts money paid to agencies, ad platforms, media and printers, not staff time. The survey publishes no advertising figure for tutoring businesses alone and does not split spending by medium, so the share spent on video is not known.

How many tutoring businesses are there in Florida and in Palm Beach, Broward and Martin counties?

County Business Patterns counted 713 exam preparation and tutoring establishments with paid employees in Florida in 2023 (Census), 7.3% of the U.S. total (our calculation). Florida’s tutoring and exam-prep establishments had revenue of $334.0 million in 2022, per the Economic Census (Census). BLS QCEW counted 6,692 private jobs in the industry in the state in 2025 (BLS). Census counted 73,482 Florida educational services businesses with no employees in 2024 (Census), the broader NAICS 611 group.

The three counties together had 184 tutoring and exam-prep establishments with employees in 2023 (our calculation), 25.8% of Florida’s (our calculation). Palm Beach and Broward held 20.2% of the state’s tutoring jobs in 2025 (our calculation); QCEW withholds Martin County’s figure to protect the confidentiality of individual employers. The three counties also had 15,284 education businesses with no employees in 2024 (our calculation), 20.8% of Florida’s total (our calculation).

How many Florida children get free books from New Worlds Reading?

New Worlds Reading enrolled 306,812 eligible Florida students in the 2024-25 school year, 28.9% of those who qualified, according to the district enrollment report the University of Florida Lastinger Center files with the Florida Department of Education (data as of June 23, 2025) (Florida DOE). The program is Florida’s state literacy initiative under Fla. Stat. 1003.485. Its purpose is to give free books to students in prekindergarten through grade 5 who read below grade level (Fla. Stat. 1003.485). It is a book-by-mail program, not a tutoring service, but it reaches the same children that reading tutors and tutoring centers serve.

The same report counted 1,060,397 eligible students out of 1,315,774 VPK-5 students in VPK programs, public schools and charter schools in 2024-25, or 80.6% (Florida DOE). Eligibility means a child is not yet reading on grade level on the state’s progress-monitoring tests. Enrolled children get 9 free books each school year, one a month from October through June, chosen by the family (New Worlds Reading).

The program has grown each year. It had 164,164 students enrolled at the end of its first year, as of June 29, 2022, when the eligibility rule was narrower (Florida DOE annual report 2021-22). Statewide enrollment of eligible students rose 15.5% from 2023-24 to 2024-25 (our calculation). By June 2025 the administrator said it had shipped 10 million books to more than 460,000 students since 2021 (New Worlds Reading). In 2024-25 alone it shipped more than 4 million books to more than 306,000 students (New Worlds Reading).

Video is written into the law. The administrator must send parents links to reading resources, including video modules, by text message and e-mail (Fla. Stat. 1003.485). The program’s family page offers short videos on phonics, vocabulary, fluency and comprehension (New Worlds Reading). Neither source says how many videos exist or how often they are watched.

In the administrator’s own survey of 21,404 New Worlds Reading families in June and July 2025, 90% said their children read or look at books more often (UF Lastinger Center annual report). The survey went to enrolled families and is not a random sample. The program also trains teachers: 12,018 literacy micro-credentials were awarded in 2024-25 (UF Lastinger Center annual report), and 702 educators from 61 counties attended its teacher professional learning (UF Lastinger Center annual report).

How many Palm Beach, Broward and Martin students are enrolled in New Worlds Reading?

Broward, Palm Beach and Martin counties together had 39,862 eligible students enrolled in New Worlds Reading in 2024-25 (our calculation from the Florida DOE district enrollment counts). That is 13.0% of statewide enrollment (our calculation). Palm Beach grew fastest of the three from 2023-24. Enrollment counts only eligible students, so a county with stronger readers has a smaller pool to enroll.

Changes are our calculation from the two district reports. The 2023-24 column comes from the program’s year-3 district report, the 2024-25 column from its year-4 report; the eligible pool is re-set each year from that year’s test data.

How is New Worlds Reading paid for, and how much goes to vendors?

The Florida DOE grant that funds New Worlds Reading at the University of Florida carries $125 million in total project dollars (125,000,000.00) for July 2021 to November 2026, according to the grant’s disbursement report certified July 21, 2025 (Florida DOE). By June 30, 2025 it had disbursed 109,866,908.38 dollars (Florida DOE). Of that, 74,653,679.07 dollars went to vendors on the budget line for book delivery, marketing and advertising (Florida DOE), or 67.9% of all spending to date (our calculation). The report does not split book delivery from marketing.

The program also takes donations through state tax credits. The law caps those credits at $60 million in each state fiscal year after 2022-23 (Fla. Stat. 1003.485). Taxpayers pledged 35,221,000 dollars under the fiscal year 2025 cap, according to the monthly pledge table filed with the Florida DOE (Florida DOE), which is 58.7% of the cap (our calculation).

How many Florida students used New Worlds Scholarship Accounts to pay for tutoring?

Step Up For Students, the nonprofit that administers the accounts, counted 24,513 students funded for New Worlds Scholarship Accounts in 2024-25 (Step Up For Students). The accounts are education savings accounts for VPK-5 public school students who struggle with reading or math. Under Fla. Stat. 1002.411 a parent may spend one on part-time tutoring services, but only from a tutor who holds a Florida teaching certificate, an adjunct certificate or a degree in the subject, has shown subject mastery, or holds a literacy micro-credential (Fla. Stat. 1002.411). Summer and after-school reading or math programs, books and curriculum also qualify.

Funded students rose about 14.0 times from 2021-22 to 2024-25 (our calculation). For 2024-25 the account was $1,200 per student, awarded first-come, first-served, according to a February 2026 Florida Senate bill analysis (Florida Senate SB 2514 analysis). In 2023-24 Step Up described the accounts as worth $500 each (Step Up For Students).

The counts do not agree. A Florida Senate analysis of April 2025, citing an e-mail from Step Up, says 31,764 students were funded in 2023-24 (Florida Senate SB 7030 analysis), against the 17,576 on Step Up’s own fact sheet. Neither source explains the gap. The Senate figures also show how little of the money reached tutors and programs: 20,268 of those students did not spend any funds, 2,045 spent all and 9,451 spent part (Florida Senate SB 7030 analysis). That is 63.8% with no spending at all (our calculation).

Broward had more New Worlds Scholarship students than any other county in 2024-25, with 3,043 (Step Up For Students). Palm Beach ranked fifth with 1,533 (Step Up For Students). Together the two counties had 4,576 students, 18.7% of the state total (our calculation). Martin County is not in Step Up’s top-10 list, so its count is not published there.

The program is now winding down. The Legislature gave it no new money in the 2025-26 budget, and Step Up says that from the 2025-26 school year no new scholarships will be awarded or funded (Step Up For Students). Families may spend leftover balances, and accounts inactive for one fiscal year close. A 2025 bill to repeal the program and a 2026 bill to reclaim balances after December 31, 2026 both died, so Fla. Stat. 1002.411 is still in the 2026 statutes.

Which other Florida scholarships can pay a private tutor?

Step Up For Students funded 265,522 students through Florida’s income-based Family Empowerment Scholarship (FES-EO) in 2024-25, as of March 2025, according to a Florida Senate analysis (Florida Senate SB 7030 analysis). It funded another 115,417 through the FES-UA scholarship for students with disabilities (Florida Senate SB 7030 analysis). That is 380,939 accounts in all (our calculation), and both kinds can pay a tutor. Fla. Stat. 1002.394 lists part-time tutoring services as an allowed expense for both (Fla. Stat. 1002.394). The tutor must hold a Florida teaching certificate or an adjunct certificate, have a degree in the subject or shown subject mastery, or be certified by a research-based training program the department approves. The Senate analysis does not say how many accounts were spent on tutoring.

A separate law covers private tutoring as a way to meet school attendance, with a certified teacher as tutor. Such a program may serve up to 25 students in a commercial building, library, church or similar facility without a special exception, rezoning or land-use change, as long as health and fire codes are met (Fla. Stat. 1002.43). The section deals with tutoring that replaces school attendance; it does not set rules for after-school tutoring centers.

How does Florida’s RAISE program use high school students as reading tutors?

Since 2021, Florida’s RAISE law has let school districts recruit high school juniors and seniors with a cumulative GPA of 3.0 or higher to tutor students in kindergarten through grade 3 (Fla. Stat. 1008.365). Tutors also need a clean suspension record, to be on track to graduate and two teacher or sponsor recommendations. Tutoring happens on school property under a district teacher’s supervision, during or after the school day.

The tutoring takes place in RAISE schools, the schools the state picks for extra reading support. The Florida Department of Education identified 1,184 of them in 2023-24, according to a 2024 Florida Senate analysis (Florida Senate SB 46 analysis). A student who gives at least 75 verified hours of tutoring is named a New Worlds Scholar and gets a pin (Fla. Stat. 1008.365). Since 2024, districts may also pay a stipend to teachers and high school students who tutor after school (Fla. Stat. 1008.365). The law sets no stipend amount, and we found no published count of RAISE tutors or New Worlds Scholars.

What do Palm Beach, Broward and Martin budget for tutoring in their 2025-26 reading plans?

Broward County Public Schools budgets $12,190,703 from the state reading allocation in its 2025-26 reading plan, with $0 on the line for tutoring programs, according to the plan it filed with the Florida Department of Education (Broward CERP 2025-26). Most of that money pays literacy coaches. The plans cover only the state Evidence-based Reading Instruction Allocation, so tutoring paid from federal or general funds does not show up in them.

Palm Beach’s plan budgets 9,163,227 dollars from the allocation, including 3,000,000 dollars for extra reading time before or after school and in summer, and leaves the tutoring line blank (Palm Beach CERP 2025-26). Martin County’s plan budgets $756,000, with $0 for tutoring (Martin CERP 2025-26).

Broward is the only one of the three that names tutoring programs in the plan. One is a one-to-one phonics program for kindergarten and first grade. The other is high-dosage, teacher-led tutoring for selected students in its most critical schools from Varsity Tutors, the tutoring brand of Nerdy Inc. The district rates Varsity Tutors as not meeting strong, moderate or promising levels of evidence and supports its use with What Works Clearinghouse practice recommendations instead (Broward CERP 2025-26). Nerdy’s own filings are covered in the filings section.

What federal rules apply to testimonials in a tutoring business’s videos?

The FTC’s Endorsement Guides, 16 CFR Part 255, apply to parent and student testimonials in tutoring ads; the current text was published on July 26, 2023 (eCFR, 16 CFR 255). The Guides are the FTC’s reading of Section 5 of the FTC Act, not a binding rule. The Guides’ one example about an education program is a learn-to-read program: a sponsored post quotes the parent of a 7-year-old but shows a child who looks 4 years old, which misrepresents the program’s effectiveness (16 CFR 255.1, Example 7).

Score and grade claims fall under the consumer endorsement rules. If a tutoring center cannot show that a featured student’s result is typical, the ad should clearly and conspicuously disclose the generally expected performance (16 CFR 255.2(b)). Videos that present people as actual consumers should use real customers in both audio and video, or say clearly that they are not (16 CFR 255.2(c)). When a claim is made in both picture and sound, as in most videos, the disclosure belongs in both the visual and audible portions (16 CFR 255.0(f)).

Connections must be disclosed too. Material connections include payment and free or discounted products, whether or not the advertiser asks for an endorsement in return (16 CFR 255.5(a)); free tutoring sessions given to a parent influencer are our example, not the Guides’. Endorsements in ads addressed to children may be of special concern because of the audience (16 CFR 255.6). The separate consumer review rule, 16 CFR Part 465, which is binding, bars giving compensation or other incentives in exchange for reviews that express a particular sentiment (16 CFR 465.4).

How many tutoring franchise centers are there, and how many are in Florida?

The nine tutoring franchise systems in the table below reported 4,029 U.S. outlets at their latest year end, the end of 2025 for seven of them and the end of 2024 for Best in Class and The Tutoring Center (our calculation from Item 20 of each franchise disclosure document, or FDD). 297 of those outlets were in Florida (our calculation), or 7.4% (our calculation). Two math and reading brands are by far the largest: Kumon and Mathnasium together held 68.4% of the outlets (our calculation).

Franchisors must give buyers an FDD, and those that register in Minnesota file it with the state’s Department of Commerce, which publishes it in its CARDS database. The outlet counts, investment ranges and fees here are what each franchisor reported in that document. Franchisors sell franchises, so they are sellers into the market. The investment range is the franchisor’s own estimate of what a buyer spends before the center opens, including fees paid to the franchisor.

Outlets are franchised and company-owned centers combined; Club Z! franchisees run in-home tutoring businesses rather than centers. Every Florida outlet in the table is franchised: Kumon’s one company-owned Florida center went to a franchisee in 2025. Club Z! is based in Tampa, and Florida holds 14.1% of its franchised businesses (our calculation). The Tutoring Center figure is its U.S. subtotal; its system total also counts centers in the United Arab Emirates. Eye Level lists no Florida centers and projects 2 openings there. Best in Class has only a marked-up redline of its 2026 FDD on CARDS, so the table uses its 2025 document. The Tutoring Center’s Minnesota registration was cancelled in June 2026, so its 2025 document is the newest. Tutor Doctor is not in the table: its newest Minnesota filing dates from 2022.

Are tutoring franchise systems growing or shrinking?

As a group they are flat: the seven tutoring franchisors with 2026 FDDs had 3,908 U.S. outlets combined at the start of 2023 and 3,919 at the end of 2025 (our calculation from their Item 20 tables). Mathnasium grew 9.7% over those three years and Kumon 3.9% (our calculation).

The other five lost a net 146 outlets (our calculation). JEI Learning Centers shrank 18.5%, Huntington Learning 15.5%, Club Z! 13.3% and Sylvan Learning 8.6% (our calculation). Most of Sylvan’s drop came in 2025, when it went from 476 centers to 433 (Sylvan FDD). Eye Level ended 2025 one outlet below where it started 2023, at 121. Best in Class went from 55 outlets at the start of 2022 to 37 at the end of 2024 (Best in Class FDD).

Huntington publishes what its centers earn: its 232 franchised centers open all of 2025 averaged $609,454 in gross revenue, with a median of $533,106 (Huntington FDD, Item 19). Those figures come from franchisee reports and are not audited. Kumon, JEI and Eye Level make no financial performance representation in Item 19.

What do tutoring franchisees pay for marketing, and what does the money buy?

Most tutoring franchisors in 2026 collect a share of each center’s sales for a shared marketing fund: Huntington franchisees pay 2% of revenue from Huntington services and eTutoring to the Huntington Advertising Fund, with a $500 monthly minimum (Huntington FDD). Club Z! takes the greater of 2% of gross sales or $50 a month (Club Z! FDD), and Best in Class currently takes 2% of gross sales, which it may raise to 3% (Best in Class FDD, 2025). Mathnasium charges a monthly marketing fee of $250 plus 2% of gross receipts and a separate digital and local marketing fee of at least $1,000 a month (Mathnasium FDD).

Sylvan charges more: 5% of gross sales to its National Advertising Fund, plus required local marketing of 6% of sales or $1,500 a month, whichever is greater (Sylvan FDD). JEI and Eye Level charge per enrolled student instead, $1 per subject a month for JEI (JEI FDD) and $1.00 per subject-student for Eye Level (Eye Level FDD). Kumon charges no ad-fund fee: it coordinates and pays for its own advertising program, though it may set up a fund later (Kumon FDD). The Tutoring Center has no national advertising fund (The Tutoring Center FDD).

Where the funds report their spending, most of the money goes into making ads, not buying space for them. Sylvan’s fund spent about 76.9% of its 2025 money on production and 13.0% on media placement (Sylvan FDD). Huntington’s fund spent 56.5% on production and 29.0% on media placement (Huntington FDD). Mathnasium’s fund put 43% into internet and digital promotion, 21% into production and 6% into TV and radio (Mathnasium FDD). The documents do not break production spending down by format, so they do not say how much went to video.

Where does video show up in tutoring franchise documents?

Mostly in teaching and training: Kumon’s 2026 FDD lets a center teach in person twice a week for about 20-30 minutes per subject, by online video conferencing, or by a hybrid of the two, and Kumon may sit in on online sessions to check them (Kumon FDD). Huntington franchisees sell eTutoring, its online tutoring service, alongside in-center tutoring, and Huntington may set rules for a teacher’s video background, equipment and dress during eTutoring (Huntington FDD). Mathnasium starts new owners on about 30 hours of online self-paced training videos, then live sessions in a virtual classroom, then about 40 more hours of videos and assignments (Mathnasium FDD).

What does Nerdy, the owner of Varsity Tutors, report about live online tutoring?

Nerdy Inc., which sells live online tutoring through Varsity Tutors, reported revenue of $179.0 million ($178,988 thousand) for 2025 in its Form 10-K (Nerdy 10-K for 2025). That is 7.5% less than the $193.4 million ($193,399 thousand) it reported for 2023 (Nerdy 10-K for 2023) (our calculation). Nerdy sells to families and to school districts, so it is a seller into the market.

Fewer families were paying, but each paid more. Nerdy had 33.2 thousand Active Members, meaning learners with a paid Learning Membership, on December 31, 2025, against 40.7 thousand at the end of 2023 (Nerdy 10-K for 2023): 18.4% fewer (our calculation). Average revenue per member per month was $364 at the end of 2025, up 21% from $302 a year earlier, after price increases for new customers (Nerdy 10-K for 2025).

Sales to schools fell faster. Institutional revenue, from Varsity Tutors for Schools products such as high-dosage tutoring and Live+AI sold to school districts, was $27.6 million ($27,607 thousand) in 2025, 22% less than in 2024 and 15% of revenue (Nerdy 10-K for 2025). The filing reports lower institutional bookings and says Nerdy scaled back its school sales effort because district funding was uncertain in 2025. Nerdy’s tutors are independent contractors it calls Experts: 15.8 thousand of them taught at least one session in 2025, down 22% from 20.2 thousand in 2024 (Nerdy 10-K for 2025). It spent $60.1 million ($60,123 thousand) on sales and marketing in 2025, 34% of revenue (Nerdy 10-K for 2025).

The sessions run on video, and the software is being rewritten with AI. Nerdy describes its rebuilt Live Learning Platform 2.0 as a collaborative workspace with 2-way video, subject-specific tools and file uploads (Nerdy 10-K for 2025). It says nearly 100% of its traffic now runs on new codebases written with AI (Nerdy 10-K for 2025), and it wrote off $7.8 million ($7,757 thousand) of older software that the new code replaced (Nerdy 10-K for 2025).

What happened to Chegg?

Chegg’s total net revenues fell to $376.9 million ($376,908 thousand) in 2025, from $617.6 million in 2024 and $716.3 million in 2023, according to its Form 10-K (Chegg 10-K for 2025): 47.4% less in two years (our calculation). Chegg sells study subscriptions to students, and its filing names Google’s AI Overviews and students’ use of generative AI as headwinds that cut its website traffic and new subscriptions.

Its 2024 annual report counted 6.6 million subscribers that year, down 14% from 7.7 million in 2023 (Chegg 10-K for 2024); the 2025 report gives no subscriber count. In 2025 Chegg cut about 640 jobs, about 56% of its workforce, citing competition and students’ adoption of generative AI (Chegg 10-K for 2025), and its sales and marketing spending fell 37% (Chegg 10-K for 2025). The online course statistics page covers Chegg alongside other e-learning companies.

How often do annual reports mention tutoring?

48 Form 10-K annual reports filed with the SEC in 2025 contain the word “tutoring”, by our count in EDGAR full-text search (the yearly counts for 2001 to 2025 are in our EDGAR counts CSV). The count rose from 23 in 2020 to 52 in 2021 and peaked at 66 in 2022 (our EDGAR count).

Most of that jump came from companies outside education. Filers outside educational services (SIC codes 8200 to 8299) went from 11 in 2020 to 40 in 2021 (our EDGAR count), while educational-services filers stayed at about a dozen a year. Some newcomers, such as Best Buy and Danaher, listed tutoring support among the benefits they added for employees during COVID-19. Blank-check companies such as Edify Acquisition list tutoring among the industries they might buy into. In 2025, educational-services companies filed 12 of the 48 (our EDGAR count).

Video comes up in many of these filings. 19 of the 2025 filings that mention tutoring also contain the word “video” (our EDGAR count), or 39.6% (our calculation). So does AI: 25 filings in 2025 contain both “tutoring” and “artificial intelligence”, and none did before 2012 (our EDGAR count). The exact phrase “online tutoring” is rare: 3 filings in 2025, from Nerdy, Stride and Zhuoxun Hongtu (our EDGAR count). A filing counts once however often it uses the word, and a mention can be brief.

What do the popular tutoring numbers really say?

Tutoring companies, college learning centers, edtech apps and the marketers who sell to them repeat four numbers more than any others: the learning pyramid’s retention rate for teaching, Bloom’s “2 sigma” for tutoring, EyeView’s conversion lift for video and the forgetting curve. We traced each one to the earliest document we could find and read what it measured. None of the four measured what it is now cited for. Pages that repeat a number are linked only as dated evidence of how it spread, never as a source for the number itself. The learning pyramid and the forgetting curve are also traced for workplace training on our training video statistics page.

“Students retain 90% of what they teach others” (the learning pyramid)

No study has ever been found behind the figure. Its best-known source, NTL Institute, answers enquiries with a letter listing retention of 90% “when they teach someone else/use immediately”, with no study, sample or date (NTL letter, archived). The letter says the pyramid was developed at NTL’s Bethel, Maine campus in the early sixties (Work-Learning Research, archived). NTL has also told enquirers it no longer has, and cannot find, “the original research that supports the numbers” (Learning Accelerators). The Internet Archive holds no learning-pyramid page on NTL’s own website.

The numbers are older than NTL’s date. Will Thalheimer’s 2006 tracing found them first printed in the training press in 1967 by D. G. Treichler, a Mobil Oil employee writing in Film and Audio-Visual Communications, who cited no research (Work-Learning Research, archived). Edgar Dale’s Cone of Experience, developed in 1946 and often shown with the percentages laid over it, had no numbers at all (Work-Learning Research, archived). Later work pushed the date further back. Thalheimer and his co-authors found the same percentages in print as early as 1914, “presented as information long known” (Work-Learning Research, archived), and a 2018 peer-reviewed search of digital libraries found versions of the pyramid in circulation for more than 160 years, concluding the models “did not originate from empirical research” (Cogent Education via ERIC). A 2014 series in Educational Technology concluded that the retention chart “cannot be supported in terms of scientific validity or logical interpretability” (Work-Learning Research, archived), and a 2007 review in the journal Education found “No specific credible research” to support it (Education via ERIC).

The top step changed as it was copied. Thalheimer found one version that raised teaching others to 95% (Work-Learning Research, archived). The tutoring form is the one Education Corner, a study-skills website, has printed since at least August 2017: students “retain abou 90% of what they’re able to teach to others”, and “the most common form of teaching others is Peer Tutoring” (Education Corner, archived). The Academic Success Center at the University of Arkansas - Fort Smith, which runs peer tutoring, copied the passage into a 2023 handout that credits the “National Training Laboratory” and says students retain nearly 90% of what they learn through teaching others (UAFS handout).

Use instead: say that tutors learn from tutoring, and cite measured effects rather than a retention percentage. Cohen, Kulik and Kulik’s 1982 meta-analysis of 65 evaluations of school tutoring programs found gains for both the students tutored and the student tutors (American Educational Research Journal via ERIC). In a 2020 experiment with 120 college students, those who taught a science text on video to a fictitious peer outscored students who restudied it on a test a week later, by d = 0.80 to 1.46 (Journal of Educational Psychology via ERIC). Neither study measured a share of material kept.

“Tutoring raises achievement by two standard deviations” (Bloom’s “2 sigma”) and “two grade levels”

The two-sigma result is real but narrow. It came from short experiments by two of Benjamin Bloom’s doctoral students, on topics new to the students, and the tutored groups got more than a tutor. In “The 2 Sigma Problem” (Educational Researcher, 1984), Bloom reported that the average tutored student scored above 98% of the students in the conventionally taught class (Bloom 1984). The evidence was the University of Chicago dissertations of Joanne Anania and Arthur J. Burke. Each sub-study ran 11 class periods (Bloom 1984). The comparison classes had about 30 students per teacher (Bloom 1984), and the tutoring condition gave a tutor to each student “or for two or three students simultaneously” (Bloom 1984). Neither Bloom’s article nor the published abstract of Anania’s study gives the total number of students.

Paul von Hippel of the University of Texas at Austin read both dissertations for a 2024 Education Next article. The students were 4th, 5th and 8th graders randomly assigned to whole-class instruction, mastery learning or tutoring, and the tutored group’s lead was measured on a post-test at the end of a three-week unit in probability or cartography (Education Next). Students in mastery learning, who got the same extra testing and feedback without a tutor, scored about 1.1 standard deviations above the whole-class group (Education Next), so much of the two sigmas came from testing and feedback. Tests written by study authors also inflate effects: Cohen, Kulik and Kulik’s 1982 meta-analysis found tutoring effects of 0.84 standard deviations on such narrow tests and 0.27 on broader standardized tests (Education Next). Bloom said he adapted his estimates from a summary by Herbert Walberg, who put tutoring at 0.40 standard deviations (Education Next).

Reviews of many studies found much smaller effects. The same Cohen, Kulik and Kulik meta-analysis put the average effect of tutoring at about 0.33 standard deviations, and only one of its 65 studies, a 1972 dissertation, reported a two-sigma result (Education Next). Kurt VanLehn’s 2011 review in Educational Psychologist tested the belief that adult human tutors reach d = 2.0 (ERIC) and found human tutoring at d = 0.79 and intelligent tutoring systems at 0.76 compared with no tutoring (ERIC). The newest meta-analysis, Nickow, Oreopoulos and Quan (2020), pooled 0.37 standard deviations across randomized trials and 0.288 in its 2024 journal version (see how much tutoring raises test scores). Von Hippel puts the typical effect of an intense, well-designed program on broad tests at about one-third of a standard deviation (Education Next).

Sellers and edtech promoters dropped the conditions. Sal Khan of Khan Academy launched its Khanmigo AI tutor with a May 2023 talk titled “The Two Sigma Solution” (Education Next). Tutors International, a private tutoring firm, wrote in March 2024 that the average student who had their own tutor “did better than 98% of students in a normal class setting”, leaving out the tests, the feedback and the small groups (Tutors International). The same page says “2-Sigma became essentially a brochure for private tutoring” (Tutors International). My Tutor Source, an online tutoring seller, added a translation Bloom never made: “2 standard deviations better means roughly converting a C grade to an A” (My Tutor Source).

“Two grade levels” claims come from sellers’ own before-and-after tests, not from comparisons. Huntington Learning Center’s location page for Plantation, in Broward County, advertised average tutoring results of 2+ grade levels of growth in reading and math, with no test, sample, hours or comparison group (Huntington, archived). Huntington’s October 2024 release on its school programs reported gains of 2.1 grade levels in reading for 4th-8th graders in a 48-hour online program over Zoom in Skokie, Illinois (Huntington via Newswire) and 1.19 grade levels in math for 6th-8th graders after 40 hours in groups of three in two Ohio districts (Huntington via Newswire). The release names no comparison group, so the gains cannot be separated from what the students would have learned in class over the same weeks.

Use instead: for an effect size, cite a pooled estimate from randomized trials with its test type, such as the 0.37 standard deviations in Nickow, Oreopoulos and Quan (2020). If Bloom is cited, say his 98% came from 11-period experiments on unfamiliar topics, with extra testing, feedback and trained tutors. For a grade-level claim, ask for the test, the hours and a comparison group.

“Video on a landing page lifts conversions by 80%”

The figure is a best case from one video vendor, EyeView, in 2009-2010, and its strongest case was a single landing page of TutorVista, an online tutoring seller. EyeView, which made the videos it tested, first reported on its blog that eToro visitors who watched its video were 20% more likely to download the product (December 2008) (EyeView, archived). In February 2009 it said UMOO viewers were 80% more likely to become depositing users, comparing viewers with non-viewers for one client (EyeView, archived). Its TutorVista case study, captured in November 2009, split new visitors from search between TutorVista’s landing page and the same page with a video that played automatically once, and reported an 86 percent increase in clicks on the Subscribe button, with no visitor counts or baseline rate (EyeView case study, archived). EyeView’s blog quoted a TutorVista manager saying the first test “boosted conversion by over 80 percent” (EyeView, archived). Its 2010 white paper said only that “in some cases” the boost was over 80% (EyeView white paper, archived), and the one test it described, eToro’s, was now almost 32 percent (EyeView white paper, archived).

The scope fell away in repeats. By August 2010 Unbounce, a landing-page software seller, wrote that video on landing pages can increase conversion by 80% (Unbounce, archived), and WordStream, a pay-per-click software seller, repeated 80% in March 2016, crediting “EyeWideDigital” (WordStream). Education marketers carried the TutorVista number itself. Elliance, a higher-education marketing agency, told colleges in April 2014 that videos on landing pages improve conversion by up to 86%, linking to an infographic rather than a study (Elliance). Beverly Boy Productions, a video producer, tells tutoring and driving schools that marketing videos can increase landing page conversions by 80%, with no source (Beverly Boy Productions).

Unbounce later checked the claim against its own customers. Its 2021 analysis of about 35,000 customer landing pages in each of three periods found video had a minimal or negative effect on conversion (Unbounce). Its April 2025 article of landing-page examples still says landing pages with video can increase conversions by 80%, linking back to its own 2010 article (Unbounce). We found no tutoring-center test behind the figure other than TutorVista’s single page in 2009.

Use instead: a center’s own split test of the same page with and without video, with visitor counts. If the EyeView figure is cited, it should read 86 percent for one TutorVista landing page in 2009, measured as Subscribe clicks and reported by the vendor that made the video.

“Students forget 70% of what they learn within 24 hours”

The number comes from Hermann Ebbinghaus’s 1885 experiments on himself with lists of nonsense syllables, and he measured the effort needed to relearn a list, not the share of a lesson forgotten. After 24 hours “about one third” of the original work was still saved when he relearned a list (Ebbinghaus 1885), and for series of 13 syllables the saving was 33.7 per cent (Ebbinghaus 1885). One hour after learning he had to spend one half of the original effort again (Ebbinghaus 1885). The “70% forgotten” form reads the relearning effort he still needed as a share of material forgotten, which he did not measure. A 2015 replication at the University of Amsterdam found the same curve, again with one subject, who spent 70 hours learning and relearning nonsense syllables (PLOS ONE).

Repeats do not agree on the 24-hour number. Lumosity, a brain-training app, says that within 24 hours “roughly 70% is gone”, citing that one-subject replication (Lumosity). The University of Minnesota’s student study-skills site says we forget about 30% of what we learned in the first 24 hours, with no source (University of Minnesota). Neither figure describes students learning school material from a tutor.

Use instead: say that memory for new material fades fastest in the first hours and days and that spaced review slows the loss, and give Ebbinghaus’s measure with its scope: about one third of the learning effort still saved after a day, for one adult learning nonsense syllables in the 1880s.

Every popular claim our research pages trace, 816 in all, is on one page: popular video statistics, fact-checked.

How we built this page

We started from the questions people search for about tutoring center video statistics and looked for the organization that actually produced each answer. Every figure on this page sits in an evidence ledger with its source, the exact table or paragraph it came from, the measurement period, the population measured, and the date we checked it. This edition holds 571 figures from 67 publishers.

On November 7, 2026, each sourced figure on this page was re-fetched with the fetch cache cleared and matched against the recorded wording: BLS occupation wage estimates and the Quarterly Census of Employment and Wages (through the BLS data API and data files) and Employment Projections (through an Internet Archive copy, because bls.gov refuses automated requests); the 2022 Economic Census, County Business Patterns, Nonemployer Statistics and the Annual Integrated Economic Survey (through the Census data API); NCES School Pulse Panel releases, Fast Facts and NAEP results, U.S. Department of Education, GAO and AmeriCorps pages (several through Internet Archive copies, because ed.gov and gao.gov refuse scripts); NBER and Annenberg EdWorkingPaper working papers, J-PAL, the National Student Support Accelerator, arXiv, ERIC records and journal articles (PMC full texts re-read through NCBI’s E-utilities when the pages answered with a bot check); survey owners’ own releases (50CAN, EdChoice, USC’s Understanding America Study, Gallup, RAND, Pew Research Center), College Board and ACT reports, and tutoring sellers’ own pages; the 2026 Florida Statutes, Florida Senate bill analyses, the University of Florida Lastinger Center, Step Up For Students and school district reading plans; eCFR for FTC rules; SEC filings on EDGAR and franchise disclosure documents filed with the Minnesota Department of Commerce; and the pages behind each popular claim, many through dated Internet Archive copies. Our YouTube figures come from the YouTube Data API and our filing counts from EDGAR full-text search. We left out Wyzant’s site and nature.com (robots rules that bar AI agents), SAGE Journals and K-12 Dive (they refuse scripts), the Princeton Review’s and Kaplan’s PDF files (their robots rules close them to every crawler), the School Pulse Panel’s spreadsheet tables and Mathnasium’s image-only revenue table (our checker cannot read them).

We only cite the producer: government agencies, regulators, company filings and official posts, and the survey owner's own report. We do not cite statistics roundups, Statista-style aggregators, or salary sites without a public method. We did not bypass paywalls or sign-up forms; where a full report was gated, we used only the figures its publisher states publicly.

Figures marked "our calculation" are arithmetic on the cited numbers, and anyone can reproduce them:

This edition contains no Mark Studios customer data.

Limitations

• Tutor counts differ by source. BLS counts 175,070 wage-and-salary tutor jobs, which leaves out self-employed tutors; its projections put the occupation at 208.4 thousand jobs with them. Census counts tutoring businesses with employees by establishment, and publishes no count of tutoring businesses without employees: its nonemployer figures cover all educational services. We print them side by side rather than pick one.
• School tutoring figures are principals’ reports. The NCES School Pulse Panel asks a sample of public school leaders what their school offered; it does not count tutors or sessions, and its newest breakdown by type of tutoring is from October 2024. No federal source separates ESSER dollars spent on tutoring.
• Effect sizes come from specific programs. Most tutoring trials tested intensive programs with trained tutors over months, often run by researchers; effects shrink in larger programs, and few trials test the weekly drop-in tutoring most private centers sell.
• Our YouTube read follows YouTube’s ranking. YouTube’s daily search limit for our key was reached after the first 100 results for ‘tutoring’ on November 7, 2026, so the sample is one search term, sorted by views, with no country filter. It is not a census of tutoring video, view counts are lifetime snapshots, and we sorted 68 channels by hand from their public text. We do not name children or small personal channels.
• Franchise and company figures are their own. Outlet counts come from each franchisor’s disclosure document filed in Minnesota, for its own fiscal year; Nerdy’s and Chegg’s figures come from their 10-K filings; tutoring sellers’ surveys and prices come from companies that sell tutoring. Each is labelled.
• Florida program counts differ by source. Step Up For Students and the Florida Senate report different counts of New Worlds Scholarship students for 2023-24; we print both.

How to cite this page

For a single figure, cite the original publisher named next to it, and link here if the verification trail is useful to your readers. Each figure has its own anchor, so you can link straight to it, for example https://www.markstudios.com/research/tutoring-center-video-statistics#DRV-210. To cite the page as a whole:

Mark Studios. "Tutoring Center Video Statistics." Last verified Nov 7, 2026. https://www.markstudios.com/research/tutoring-center-video-statistics

@misc{markstudios_tutoring_center_video_statistics_2026,
  author = {{Mark Studios}},
  title = {{Tutoring Center Video Statistics}},
  year = {2026},
  month = nov,
  url = {https://www.markstudios.com/research/tutoring-center-video-statistics},
  note = {Last verified 2026-11-07}
}

Our calculations and the evidence table are free to reuse with attribution and a link to this page. Each underlying figure remains the claim of its original publisher, and some publishers ask for a link when their figures are reused.

Sources

Download the evidence table (CSV)

1. National Bureau of Economic Research (Nickow, Oreopoulos & Quan), The Impressive Effects of Tutoring on PreK-12 Learning: A Systematic Review and Meta-Analysis of the Experimental Evidence (NBER Working Paper 27476) (2020-07). Accessed 2026-11-07.
2. National Bureau of Economic Research (Nickow, Oreopoulos & Quan), The Impressive Effects of Tutoring on PreK-12 Learning: A Systematic Review and Meta-Analysis of the Experimental Evidence (NBER Working Paper 27476) (2020-07). Accessed 2026-11-07.
3. American Educational Research Journal (Nickow, Oreopoulos & Quan), abstract indexed by ERIC (Institute of Education Sciences), The Promise of Tutoring for PreK-12 Learning: A Systematic Review and Meta-Analysis of the Experimental Evidence (American Educational Research Journal, v61 n1 p74-107, 2024) (2024). Accessed 2026-11-07.
4. Abdul Latif Jameel Poverty Action Lab (J-PAL), MIT, The Transformative Potential of Tutoring for Pre K-12 Learning Outcomes: Lessons from Randomized Evaluations (2020-09-16). Accessed 2026-11-07.
5. Annenberg Institute at Brown University, EdWorkingPaper 24-1031 (Kraft, Schueler & Falken), What Impacts Should We Expect from Tutoring at Scale? Exploring Meta-Analytic Generalizability (EdWorkingPaper 24-1031) (2024-10). Accessed 2026-11-07.
6. Annenberg Institute at Brown University, EdWorkingPaper 24-1031 (Kraft, Schueler & Falken), What Impacts Should We Expect from Tutoring at Scale? Exploring Meta-Analytic Generalizability (EdWorkingPaper 24-1031) (2024-10). Accessed 2026-11-07.
7. National Bureau of Economic Research (Guryan, Ludwig, Bhatt, Cook et al.), Not Too Late: Improving Academic Outcomes Among Adolescents (NBER Working Paper 28531) (2021-03). Accessed 2026-11-07.
8. National Bureau of Economic Research (Bhatt, Guryan, Khan, LaForest-Tucker & Mishra), Can Technology Facilitate Scale? Evidence from a Randomized Evaluation of High Dosage Tutoring (NBER Working Paper 32510) (2024-05). Accessed 2026-11-07.
9. Annenberg Institute at Brown University, EdWorkingPaper 20-335 (Kraft & Falken), A Blueprint for Scaling Tutoring Across Public Schools (EdWorkingPaper 20-335) (2021-01). Accessed 2026-11-07.
10. Abdul Latif Jameel Poverty Action Lab (J-PAL), summary of Carlana & La Ferrara's evaluation, Improving Educational Outcomes through Online Tutoring during Schools Closures in Italy during the Covid-19 Outbreak (n.d. (timeline 2020-2022)). Accessed 2026-11-07.
11. Annenberg Institute at Brown University, EdWorkingPaper 21-350 (Carlana & La Ferrara), Apart but Connected: Online Tutoring and Student Outcomes during the COVID-19 Pandemic (EdWorkingPaper 21-350) (2021-02). Accessed 2026-11-07.
12. Annenberg Institute at Brown University, EdWorkingPaper 23-743 (Gortazar, Hupkau & Roldán), Online tutoring works: Experimental evidence from a program with vulnerable children (EdWorkingPaper 23-743) (2023-03). Accessed 2026-11-07.
13. Annenberg Institute at Brown University, EdWorkingPaper 23-743 (Gortazar, Hupkau & Roldán), Online tutoring works: Experimental evidence from a program with vulnerable children (EdWorkingPaper 23-743) (2023-03). Accessed 2026-11-07.
14. Annenberg Institute at Brown University, EdWorkingPaper 26-1496 (Calle, Gortazar, Hernandez-de-Benito, Hupkau, Molina-Millán & Roldán-Monés), From Pilot to Policy: Experimental Evidence from Scaling Online Tutoring (EdWorkingPaper 26-1496) (2026-06). Accessed 2026-11-07.
15. Annenberg Institute at Brown University, EdWorkingPaper 26-1496 (Calle, Gortazar, Hernandez-de-Benito, Hupkau, Molina-Millán & Roldán-Monés), From Pilot to Policy: Experimental Evidence from Scaling Online Tutoring (EdWorkingPaper 26-1496) (2026-06). Accessed 2026-11-07.
16. Annenberg Institute at Brown University, EdWorkingPaper 22-568 (Kraft, List, Livingston & Sadoff), Online Tutoring by College Volunteers: Experimental Evidence from a Pilot Program (EdWorkingPaper 22-568) (2022-04). Accessed 2026-11-07.
17. Annenberg Institute at Brown University, EdWorkingPaper 22-568 (Kraft, List, Livingston & Sadoff), Online Tutoring by College Volunteers: Experimental Evidence from a Pilot Program (EdWorkingPaper 22-568) (2022-04). Accessed 2026-11-07.
18. Annenberg Institute at Brown University, EdWorkingPaper 24-955 (Robinson, Pollard, Novicoff, White & Loeb), The Effects of Virtual Tutoring on Young Readers: Results from a Randomized Controlled Trial (EdWorkingPaper 24-955) (2024-11). Accessed 2026-11-07.
19. Annenberg Institute at Brown University, EdWorkingPaper 24-955 (Robinson, Pollard, Novicoff, White & Loeb), The Effects of Virtual Tutoring on Young Readers: Results from a Randomized Controlled Trial (EdWorkingPaper 24-955) (2024-11). Accessed 2026-11-07.
20. Annenberg Institute at Brown University, EdWorkingPaper 25-1176 (Hashim, Pace Miles & Croke), Experimental Evidence on the Impact of Tutoring Format and Tutors: Findings from an Early Literacy Tutoring Program (EdWorkingPaper 25-1176) (2025-04). Accessed 2026-11-07.
21. National Student Support Accelerator (Stanford University SCALE Initiative), High-Impact Tutoring Database (live database (read 2026-11-07)). Accessed 2026-11-07.
22. Stanford University researchers (Wang, Ribeiro, Robinson, Loeb & Demszky), arXiv preprint, Tutor CoPilot: A Human-AI Approach for Scaling Real-Time Expertise (arXiv:2410.03017) (2024-10-03). Accessed 2026-11-07.
23. Annenberg Institute at Brown University, EdWorkingPaper 26-1551 (Oreopoulos & Low), One Click Away: AI Tutoring with Khanmigo in a Two-Year School Experiment (EdWorkingPaper 26-1551) (2026-08). Accessed 2026-11-07.
24. Proceedings of the National Academy of Sciences (Bastani, Bastani, Sungu, Ge, Kabakcı & Mariman), Generative AI without guardrails can harm learning: Evidence from high school mathematics (PNAS 122(26), 2025) (2025-06-25). Accessed 2026-11-07.
25. Annenberg Institute at Brown University, EdWorkingPaper 26-1598 (Liu, Sweet, Chen et al.), The Effects of Course-Integrated AI Tutoring on Student Performance and Engagement: A Randomized University Trial (EdWorkingPaper 26-1598) (2026-10). Accessed 2026-11-07.
26. Scientific Reports (Kestin, Miller, Klales, Milbourne & Ponti, Harvard University), AI tutoring outperforms in-class active learning: an RCT introducing a novel research-based design in an authentic educational setting (Scientific Reports 15:17458, 2025) (2025-06-03). Accessed 2026-11-07.
27. Noetel et al., Review of Educational Research (American Educational Research Association), Video Improves Learning in Higher Education: A Systematic Review (2021-02-17). Accessed 2026-11-07.
28. Annenberg Institute at Brown University, EdWorkingPaper 21-367 (Kim, Goodman & West), Kumon In: The Recent, Rapid Rise of Private Tutoring Centers (EdWorkingPaper 21-367) (2024-07). Accessed 2026-11-07.
29. Annenberg Institute at Brown University, EdWorkingPaper 26-1619 (Chen), The Returns to Private Tutoring Centers (EdWorkingPaper 26-1619) (2026-10). Accessed 2026-11-07.
30. 50CAN (survey by Edge Research), The State of Educational Opportunity in America: 2nd Edition (50CAN Education Opportunity Survey) (2026-02). Accessed 2026-11-07.
31. EdChoice (poll by Morning Consult), A Kaleidoscope View of K-12 Tutoring in America (2025-11-17). Accessed 2026-11-07.
32. USC Center for Economic and Social Research (Understanding America Study), Two Percent of U.S. Children Receive High Quality Tutoring, Despite Billions Funneled into School Systems (2023-03-10). Accessed 2026-11-07.
33. 50CAN (survey by Edge Research), Florida: 50CAN Education Opportunity Survey (state report, 2024) (2024-10). Accessed 2026-11-07.
34. Gallup and Learning Heroes, New Data Find Nearly Nine in 10 Parents Say Their Child Is at or Above Grade Level in Reading and Math, Despite Historic Declines (press release for B-flation: How Good Grades Can Sideline Parents) (2023-11-15). Accessed 2026-11-07.
35. Sylvan Learning (seller of tutoring), How much does Sylvan cost? (Sylvan Learning pricing page) (undated). Accessed 2026-11-07.
36. Preply (seller: online tutoring marketplace), Online math tutors (Preply) (undated). Accessed 2026-11-07.
37. College Board, 2025 SAT Suite of Assessments Annual Report: Total Group (2025-09-30). Accessed 2026-11-07.
38. College Board, SAT Participation in the Class of 2025 Surpasses 2 Million Test Takers for the First Time Since 2020 (2025-09-30). Accessed 2026-11-07.
39. ACT, The ACT Profile Report: National, Graduating Class 2025 (2025-10-15). Accessed 2026-11-07.
40. ACT, ACT Releases 2025 Grad Class Data on College and Career Readiness (press release) (2025-10-15). Accessed 2026-11-07.
41. College Board, 2025 SAT Suite of Assessments Annual Report: Florida (2025-09-30). Accessed 2026-11-07.
42. ACT, Average ACT Test Scores by State, Graduating Class of 2025 (percent meeting benchmarks) (2025-10-15). Accessed 2026-11-07.
43. Kaplan (seller of SAT/ACT prep), 2026-2027 College Admissions Trends (Kaplan Admissions Officer Survey) (2026-10-13). Accessed 2026-11-07.
44. The Princeton Review (seller of SAT/ACT prep and tutoring), The Princeton Review 2026 College Hopes & Worries Survey Findings (press release) (2026-03-10). Accessed 2026-11-07.
45. Huntington Learning Center (seller of tutoring and test prep; survey by YouGov), Parents Express Uncertainty Over College Future as Nearly All Agree Supplemental Education Is Key to Post-High School Success (Huntington Learning Center press release) (2025-05-22). Accessed 2026-11-07.
46. Huntington Learning Center (seller of tutoring and test prep), Huntington Learning Center - Personalized Tutoring & Test Prep (home page) (2026-10-12). Accessed 2026-11-07.
47. Kumon Institute of Education (seller of after-school tutoring), Company Profile | Kumon Group (2026-06). Accessed 2026-11-07.
48. Kumon North America (seller of after-school tutoring), Kumon Press Kit: Fact Sheet (2014-12). Accessed 2026-11-07.
49. Varsity Tutors, a Nerdy company (seller of online tutoring), Varsity Tutors home page (undated). Accessed 2026-11-07.
50. YouTube (Google), The global campus: How YouTube is empowering learners everywhere (2026-04-28). Accessed 2026-11-07.
51. Google (Think with Google), Self-directed learning from YouTube (2017-12). Accessed 2026-11-07.
52. Khan Academy (with College Board), College Board & Khan Academy Mark 10 Million Milestone of Free Official SAT Practice (2020-01-27). Accessed 2026-11-07.
53. Pew Research Center, About a quarter of U.S. teens have used ChatGPT for schoolwork, double the share in 2023 (2025-01-15). Accessed 2026-11-07.
54. Pew Research Center, How Teens Use and View AI (2026-02-24). Accessed 2026-11-07.
55. Gallup and Walton Family Foundation, Gen Z's AI Adoption Steady, but Skepticism Climbs (2026-04-08). Accessed 2026-11-07.
56. RAND Corporation (American School District Panel), Districts' Pandemic Recovery Efforts as COVID-Relief Aid Expires: Selected Findings from the Spring 2024 American School District Panel Survey (RR-A956-23) (2024-09). Accessed 2026-11-07.
57. U.S. Department of Education, National Center for Education Statistics (NCES), Public School Leaders Reports Suggest No Overall Change in the Number of Teaching Positions Since Last School Year (press release) (2024-12-12). Accessed 2026-11-07.
58. U.S. Department of Education, National Center for Education Statistics (NCES), Forty-Four Percent of Public School Students Began 2023-24 Year Behind Grade Level in at Least One Academic Subject, Principals Say (press release) (2023-12-14). Accessed 2026-11-07.
59. U.S. Department of Education, National Center for Education Statistics (NCES), School Pulse Panel (home page) (2025-09-23). Accessed 2026-11-07.
60. U.S. Department of Education, National Center for Education Statistics (NCES), Administrators Report Roughly Half of Public School Students Began 2022-23 School Year Behind Grade Level in At Least One Academic Subject (press release) (2023-02-09). Accessed 2026-11-07.
61. U.S. Department of Education, National Center for Education Statistics (NCES), Fast Facts: Back-to-school statistics (2024). Accessed 2026-11-07.
62. U.S. Government Accountability Office (GAO), Pandemic Learning: As Students Struggled to Learn, Teachers Reported Few Strategies as Particularly Helpful to Mitigate Learning Loss (GAO-22-104487) (2022-05-10). Accessed 2026-11-07.
63. U.S. Government Accountability Office (GAO), K-12 Education: School Districts Reported Spending Initial COVID Relief Funds on Meeting Students' Needs and Continuing School Operations (GAO-24-106913) (2024-09-23). Accessed 2026-11-07.
64. U.S. Department of Education (Federal Register), American Rescue Plan Act Elementary and Secondary School Emergency Relief Fund (interim final requirements), 86 FR 21195 (2021-04-22). Accessed 2026-11-07.
65. U.S. Department of Education, U.S. Department of Education Answers President Biden's Call to Action to Spur Academic Recovery (press release, July 5, 2022) (2022-07-05). Accessed 2026-11-07.
66. AmeriCorps, AmeriCorps joins White House and Department of Education to Launch National Effort to Support Student Success (July 5, 2022) (2022-07-05). Accessed 2026-11-07.
67. AmeriCorps, AmeriCorps and Partners Surpass President Biden's Tutor and Mentor Recruitment Goal a Year Ahead of Schedule (October 10, 2024) (2024-10-10). Accessed 2026-11-07.
68. U.S. Department of Education, National Center for Education Statistics (NCES), NAEP, Highlights from the 2022 NAEP Long-Term Trend Reading and Mathematics Assessments at Age 9 (Nation's Report Card) (2022-09-01). Accessed 2026-11-07.
69. U.S. Department of Education, National Center for Education Statistics (NCES), NAEP, Highlights from the 2023 NAEP Long-Term Trend Assessments at Age 13 (Nation's Report Card) (2023-06-21). Accessed 2026-11-07.
70. U.S. Department of Education, National Center for Education Statistics (NCES), NAEP, 2024 NAEP Mathematics Assessment: Results at Grades 4 and 8 (Nation's Report Card) (2025-01-29). Accessed 2026-11-07.
71. U.S. Department of Education, National Center for Education Statistics (NCES), NAEP, 2024 NAEP Reading Assessment: Results at Grades 4 and 8 (Nation's Report Card) (2025-01-29). Accessed 2026-11-07.
72. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
73. U.S. Bureau of Labor Statistics, Table 1.2 Occupational projections, 2025–35, and worker characteristics, 2025 (2026-08-27). Accessed 2026-11-07.
74. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
75. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
76. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
77. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
78. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
79. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
80. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
81. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
82. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
83. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
84. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
85. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
86. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
87. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
88. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
89. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
90. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
91. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
92. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2025 estimates (2026-05-15). Accessed 2026-11-07.
93. U.S. Bureau of Labor Statistics, QCEW Open Data: 2025 annual averages, industry 611691 (CSV) (2026-08-21). Accessed 2026-11-07.
94. U.S. Bureau of Labor Statistics, QCEW Open Data: 2019 annual averages, industry 611691 (CSV) (2020-08-28). Accessed 2026-11-07.
95. U.S. Census Bureau, County Business Patterns: 2023 (Census Data API, dataset 2023/cbp) (2025-06-26). Accessed 2026-11-07.
96. U.S. Census Bureau, 2022 Economic Census: Summary Statistics for the U.S., States, and Selected Geographies (Census Data API, dataset 2022/ecnbasic) (undated). Accessed 2026-11-07.
97. U.S. Census Bureau, Annual Integrated Economic Survey 2024: Summary Statistics for Employer Firms in the U.S. (Census Data API, dataset 2024/aiesbasic) (2026). Accessed 2026-11-07.
98. U.S. Census Bureau, 2022 Economic Census: Industry by Products Statistics for the U.S. (Census Data API, dataset 2022/ecnnapcsind) (undated). Accessed 2026-11-07.
99. U.S. Census Bureau, Nonemployer Statistics: 2024 (Census Data API, dataset 2024/nonemp) (2026). Accessed 2026-11-07.
100. U.S. Census Bureau, Nonemployer Statistics: 2023 (Census Data API, dataset 2023/nonemp) (2025). Accessed 2026-11-07.
101. U.S. Census Bureau, County Business Patterns: 2023 (Census Data API, dataset 2023/cbp) (2025-06-26). Accessed 2026-11-07.
102. U.S. Census Bureau, Annual Integrated Economic Survey: Selected Sectors: Detailed Operating Expenses for Employer Firms in the U.S.: 2024 (Census Data API, dataset 2024/aiesexp02) (2026). Accessed 2026-11-07.
103. U.S. Census Bureau, County Business Patterns: 2023 (Census Data API, dataset 2023/cbp) (2025-06-26). Accessed 2026-11-07.
104. U.S. Census Bureau, County Business Patterns: 2023 (Census Data API, dataset 2023/cbp) (2025-06-26). Accessed 2026-11-07.
105. Florida Department of Education / University of Florida Lastinger Center for Learning, New Worlds Reading Initiative Year 4 District Enrollment, 2024-25 (Florida Department of Education) (2025-06-23). Accessed 2026-11-07.
106. Florida Legislature (Florida Statutes, via The Florida Senate), Florida Statutes, Section 1003.485 (2026): The New Worlds Reading Initiative (2026). Accessed 2026-11-07.
107. University of Florida Lastinger Center for Learning (New Worlds Reading), Free Books for Florida Students | New Worlds Reading (2026). Accessed 2026-11-07.
108. Florida Department of Education / University of Florida Lastinger Center for Learning, New Worlds Reading Initiative Annual Report 2021-22 (Florida Department of Education) (2022-06-29). Accessed 2026-11-07.
109. University of Florida Lastinger Center for Learning (New Worlds Reading), New Worlds Reading Initiative Celebrates Major Milestone: 10 Million Books Delivered to Florida Families (2025-06-18). Accessed 2026-11-07.
110. University of Florida Lastinger Center for Learning (New Worlds Reading), Data Shows How New Worlds Reading Makes a Difference (2026-04-17). Accessed 2026-11-07.
111. University of Florida Lastinger Center for Learning (New Worlds Reading), Family Resources | New Worlds Reading (2026). Accessed 2026-11-07.
112. University of Florida Lastinger Center for Learning, New Worlds Reading Initiative 2024-2025 Annual Enrollment Report (2025). Accessed 2026-11-07.
113. Florida Department of Education / University of Florida Lastinger Center for Learning, New Worlds Reading Initiative Year 3 District Enrollment, 2023-24 (Florida Department of Education) (2024-07-02). Accessed 2026-11-07.
114. Florida Department of Education, New Worlds Reading Initiative Project Disbursement Report (DOE 399), grant 9337C, as of June 30, 2025 (2025-07-21). Accessed 2026-11-07.
115. Florida Department of Education / University of Florida Lastinger Center for Learning, New Worlds Reading Monthly Contributions: Tax Credit (FY24-FY26 cap years) (2025). Accessed 2026-11-07.
116. Step Up For Students (state-approved scholarship funding organization; administers the program), New Worlds Scholarships fact sheet (February 2025) (2025-02). Accessed 2026-11-07.
117. Florida Legislature (Florida Statutes, via The Florida Senate), Florida Statutes, Section 1002.411 (2026): New Worlds Scholarship Accounts (2026). Accessed 2026-11-07.
118. Step Up For Students (state-approved scholarship funding organization; administers the program), New Worlds Reading Scholarships fact sheet (2023-24 Fast Facts) (2024-05). Accessed 2026-11-07.
119. The Florida Senate, Committee on Appropriations (bill analysis), Bill Analysis and Fiscal Impact Statement, SB 2514 (Education Pre-K-12), February 19, 2026 (2026-02-19). Accessed 2026-11-07.
120. The Florida Senate, Committee on Appropriations (bill analysis), Bill Analysis and Fiscal Impact Statement, CS/SB 7030 (Educational Scholarship Programs), April 3, 2025 (2025-04-03). Accessed 2026-11-07.
121. Step Up For Students (state-approved scholarship funding organization; administers the program), New Worlds Scholarship Accounts Program Parent & Guardian Handbook (2026-2027) (2026-07-23). Accessed 2026-11-07.
122. Florida Legislature (Florida Statutes, via The Florida Senate), Florida Statutes, Section 1002.394 (2026): The Family Empowerment Scholarship Program (2026). Accessed 2026-11-07.
123. Florida Legislature (Florida Statutes, via The Florida Senate), Florida Statutes, Section 1002.43 (2026): Private tutoring programs (2026). Accessed 2026-11-07.
124. Florida Legislature (Florida Statutes, via The Florida Senate), Florida Statutes, Section 1008.365 (2026): Reading Achievement Initiative for Scholastic Excellence Act (2026). Accessed 2026-11-07.
125. The Florida Senate, Committee on Rules (bill analysis), Bill Analysis and Fiscal Impact Statement, SB 46 (Reading Achievement Initiative for Scholastic Excellence Program), 2024 (2024-01-23). Accessed 2026-11-07.
126. Broward County Public Schools (CERP filed with the Florida Department of Education), Broward County Public Schools CERP for 2025-2026 (2025). Accessed 2026-11-07.
127. School District of Palm Beach County (CERP filed with the Florida Department of Education), School District of Palm Beach County CERP for 2025-2026 (2025). Accessed 2026-11-07.
128. Martin County School District (CERP filed with the Florida Department of Education), Martin County School District CERP for 2025-2026 (2025). Accessed 2026-11-07.
129. Federal Trade Commission (rule text via eCFR, Office of the Federal Register), 16 CFR Part 255: Guides Concerning Use of Endorsements and Testimonials in Advertising (2023-07-26). Accessed 2026-11-07.
130. Federal Trade Commission (rule text via eCFR, Office of the Federal Register), 16 CFR Part 465: Use of Consumer Reviews and Testimonials (2024-08-22). Accessed 2026-11-07.
131. Kumon North America, Inc. (franchise disclosure document filed with the Minnesota Department of Commerce), Kumon North America, Inc. Franchise Disclosure Document, Clean FDD 2026 (issuance date March 27, 2026), Minnesota CARDS file 2869 (document 35725-202603-05) (2026-03-27). Accessed 2026-11-07.
132. Mathnasium Franchisor LLC (franchise disclosure document filed with the Minnesota Department of Commerce), Mathnasium Franchisor LLC Franchise Disclosure Document, Clean FDD 2026 (issuance date April 30, 2026), Minnesota CARDS file 11343 (document 36620-202605-05) (2026-04-30). Accessed 2026-11-07.
133. Sylvan Learning, LLC (franchise disclosure document filed with the Minnesota Department of Commerce), Sylvan Learning, LLC Franchise Disclosure Document, Clean FDD 2026 (issuance date April 24, 2026), Minnesota CARDS file 7878 (document 36207-202604-03) (2026-04-24). Accessed 2026-11-07.
134. Club Z!, Inc. (franchise disclosure document filed with the Minnesota Department of Commerce), Club Z!, Inc. Franchise Disclosure Document, Clean FDD 2026 (issuance date April 30, 2026), Minnesota CARDS file 9886 (document 36258-202604-08) (2026-04-30). Accessed 2026-11-07.
135. Huntington Learning Centers, Inc. (franchise disclosure document filed with the Minnesota Department of Commerce), Huntington Learning Centers, Inc. Franchise Disclosure Document, Clean FDD 2026 (issuance date April 16, 2026), Minnesota CARDS file 2778 (document 36163-202604-10) (2026-04-16). Accessed 2026-11-07.
136. Daekyo America, Inc. (Eye Level Learning Centers; franchise disclosure document filed with the Minnesota Department of Commerce), Daekyo America, Inc. (Eye Level Learning Center) Franchise Disclosure Document, Clean FDD 2026 (issuance date April 8, 2026), Minnesota CARDS file 7632 (document 36078-202604-04) (2026-04-08). Accessed 2026-11-07.
137. The Tutoring Center Franchise Corp. (franchise disclosure document filed with the Minnesota Department of Commerce), The Tutoring Center Franchise Corp. Franchise Disclosure Document, Clean FDD 2025 (issuance date March 21, 2025), Minnesota CARDS file 6718 (document 33052-202503-07) (2025-03-21). Accessed 2026-11-07.
138. JEI Learning Centers, LLC (franchise disclosure document filed with the Minnesota Department of Commerce), JEI Learning Centers, LLC Franchise Disclosure Document, Clean FDD 2026 'B' (date of issuance April 14, 2026), Minnesota CARDS file 10436 (document 36013-202604-12) (2026-04-14). Accessed 2026-11-07.
139. BiC Franchise System Corporation (Best in Class Education Center; franchise disclosure document filed with the Minnesota Department of Commerce), BiC Franchise System Corporation (Best in Class Education Center) Franchise Disclosure Document, Clean FDD 2025 (issuance date May 27, 2025), Minnesota CARDS file 10758 (document 34289-202505-07) (2025-05-27). Accessed 2026-11-07.
140. Nerdy Inc. (Varsity Tutors; SEC Form 10-K), Nerdy Inc. Annual Report on Form 10-K for the fiscal year ended December 31, 2025 (2026-02-26). Accessed 2026-11-07.
141. Nerdy Inc. (Varsity Tutors; SEC Form 10-K), Nerdy Inc. Annual Report on Form 10-K for the fiscal year ended December 31, 2023 (2024-02-27). Accessed 2026-11-07.
142. Chegg, Inc., Chegg, Inc. Annual Report on Form 10-K for the fiscal year ended December 31, 2025 (2026-03-09). Accessed 2026-11-07.
143. Chegg, Inc., Chegg, Inc. Annual Report on Form 10-K for the fiscal year ended December 31, 2024 (2025-02-24). Accessed 2026-11-07.
144. NTL Institute for Applied Behavioral Science (letter hosted by Work-Learning Research), Thanks for your inquiry of NTL Institute (NTL's learning pyramid letter) (2006-11-03). Accessed 2026-11-07.
145. Work-Learning Research (Will Thalheimer), quoting NTL Institute, NTL continues its delusions (2006-11-03). Accessed 2026-11-07.
146. Learning Accelerators (Adrian Snook), Moonrakers! - Learning Accelerators (2019-12-05). Accessed 2026-11-07.
147. Work-Learning Research (Will Thalheimer), People remember 10%, 20%...Oh Really? (2006-05-01). Accessed 2026-11-07.
148. Work-Learning Research (Will Thalheimer), Mythical Retention Data & The Corrupted Cone (2015-01-05). Accessed 2026-11-07.
149. Kåre Letrud and Sigbjørn Hernes, Cogent Education (record via ERIC, U.S. Department of Education), Excavating the Origins of the Learning Pyramid Myths (2018-01-01). Accessed 2026-11-07.
150. James P. Lalley and Robert H. Miller, Education v128 n1 (record via ERIC, U.S. Department of Education), The Learning Pyramid: Does It Point Teachers in the Right Direction? (2007-09-01). Accessed 2026-11-07.
151. Education Corner (education resources website; repeater), The Learning Pyramid (2017-08-13). Accessed 2026-11-07.
152. University of Arkansas - Fort Smith, Academic Success Center (college tutoring center; repeater), The Learning Pyramid (handout) (2023-05-03). Accessed 2026-11-07.
153. American Educational Research Journal (Cohen, Kulik and Kulik; record via ERIC, U.S. Department of Education), Educational Outcomes of Tutoring: A Meta-analysis of Findings (1982-06-01). Accessed 2026-11-07.
154. Journal of Educational Psychology (Fiorella and Kuhlmann; record via ERIC, U.S. Department of Education), Creating Drawings Enhances Learning by Teaching (2020-05-01). Accessed 2026-11-07.
155. Benjamin S. Bloom, Educational Researcher 13(6) (copy hosted by gwern.net), The 2 Sigma Problem: The Search for Methods of Group Instruction as Effective as One-to-One Tutoring (1984-06-01). Accessed 2026-11-07.
156. Paul T. von Hippel (University of Texas at Austin), Education Next 24(2), Two-Sigma Tutoring: Separating Science Fiction from Science Fact (2024-03-07). Accessed 2026-11-07.
157. Kurt VanLehn, Educational Psychologist 46(4) (record via ERIC, U.S. Department of Education), The Relative Effectiveness of Human Tutoring, Intelligent Tutoring Systems, and Other Tutoring Systems (2011-10-01). Accessed 2026-11-07.
158. Tutors International (private tutoring seller; repeater), Why the 2-Sigma Experiment is the Best Advocate for Private Tutoring (2024-03-18). Accessed 2026-11-07.
159. My Tutor Source (online tutoring seller; repeater), Bloom's 2 Sigma Problem: A Summary (2022-08-20). Accessed 2026-11-07.
160. Huntington Learning Center (tutoring and test prep franchisor; seller), Huntington Learning Center - Plantation, FL (2024-10-09). Accessed 2026-11-07.
161. Huntington Learning Center (tutoring and test prep franchisor; press release via Newswire), Huntington Learning Center Delivers Transformative Results With High-Dosage Tutoring Nationwide (2024-10-15). Accessed 2026-11-07.
162. EyeView Digital (video production and testing vendor; sold the videos it tested), EyeView - The new "Conversion" game (2008-12-04). Accessed 2026-11-07.
163. EyeView Digital (video production and testing vendor; sold the videos it tested), UMOO - a prime beef example of EyeView's success (2009-02-10). Accessed 2026-11-07.
164. EyeView Digital (video production and testing vendor; sold the video it tested), EyeView case study: TutorVista (2009-11-08). Accessed 2026-11-07.
165. EyeView Digital (video production and testing vendor; sold the videos it tested), Another video conversion success story (2009-11-10). Accessed 2026-11-07.
166. EyeView Digital (video production and testing vendor; sold the videos it tested), Making Video Accountable (EyeView white paper) (2010). Accessed 2026-11-07.
167. Unbounce (landing-page software seller), The benefits of using video on landing pages (2010-08). Accessed 2026-11-07.
168. WordStream (PPC advertising software seller), 6 Ways to Use Video to Increase Conversion Rates (2016-03-30). Accessed 2026-11-07.
169. Elliance (higher-education marketing agency; repeater), Making a case for video in higher education marketing (2014-04-18). Accessed 2026-11-07.
170. Beverly Boy Productions (video production seller; repeater), Video Production for Tutoring & Driving Schools (2022-07-29). Accessed 2026-11-07.
171. Unbounce (landing-page software seller; analysis of its own customers' pages), Video on landing pages means more conversions, right? Wrong. Here's why (2021-08-26). Accessed 2026-11-07.
172. Unbounce (landing-page software seller; repeater), 15 high-converting landing page examples (+ why they work) (2025-04-06). Accessed 2026-11-07.
173. Hermann Ebbinghaus (via Classics in the History of Psychology, York University), Memory: A Contribution to Experimental Psychology, Chapter 7 (1885). Accessed 2026-11-07.
174. Murre and Dros, PLOS ONE (University of Amsterdam), Replication and Analysis of Ebbinghaus' Forgetting Curve (2015-07-06). Accessed 2026-11-07.
175. Lumosity (Lumos Labs, brain-training app seller; repeater), The Forgetting Curve (2026). Accessed 2026-11-07.
176. University of Minnesota, EffectiveU (student study-skills site; repeater), Curve of Forgetting (2026). Accessed 2026-11-07.

Update policy and change log

We re-verify every figure quarterly and when a major source publishes a new edition. The next scheduled review is February 2027. Corrections go to contact@markstudios.com.

• Nov 7, 2026: First published with 571 figures from 67 publishers, each checked against its original source.