Public speaking statistics: what the primary sources actually say

Disclosure: NOXI publishes this page and is the presentation tool named in its slide-building sections. Every statistic on the page comes from the primary study or official survey linked beside it, each opened on 14 October 2026; nothing here was commissioned or funded by those sources.

Most pages that rank for this query repeat "77% of people fear public speaking" and "it is the number one fear, worse than death" without naming a study. This page does the opposite: every number links to the survey or paper it comes from, with the question wording, the sample, the year and the margin of error, and every popular claim is traced back to where it started. It is a dated edition, built to be refreshed when Chapman publishes its 2026 wave.

Public speaking statistics at a glance (checked 14 October 2026)

The ten numbers below are the ones worth citing about public speaking, fear and attention; each comes from a named survey or study with a known sample and year. Use the "what it measures" column before you quote any of them: most errors in this topic come from quoting a number for something it never measured.

How many people are afraid of public speaking?

In 2025, 33.7% of US adults said they were afraid or very afraid of public speaking, according to the Chapman University Survey of American Fears; in 2024 the figure was 29.0% and in 2023 it was 28.7%. Counting everyone who reported at least slight fear, the 2025 share is 68.5%, so roughly two in three adults feel some fear and one in three feel a strong one.

The Chapman survey is the best public source for a current US number because it asks the same question every year of a probability-based national sample. In 2025 the survey firm SSRS fielded it online to 1,015 adults from its Opinion Panel between 24 March and 8 April 2025; the panel is recruited from a random sample of US postal addresses and prepaid cell numbers, and results are weighted to the US adult population. The question reads: "How afraid are you of the following?", with public speaking as one item and four answers: very afraid, afraid, slightly afraid, not afraid (Wave 11 methods report). Chapman's headline figure combines the first two answers.

Headline shares are as Chapman publishes them in its lists for 2023, 2024 and 2025. The answer splits come from the weighted frequency tables in each year's methods report (2023, 2024, 2025); in 2024 the two rounded parts add to 29.1% while the list reports 29.0%, and in 2025 they add to 33.6% while the list reports 33.7%, both rounding differences. The 2023 rank is Chapman's own numbering, the 2025 rank is stated in Chapman's key findings, and the 2024 rank is the position in Chapman's list sorted from highest to lowest. Ranks are not comparable across years because the list of fears changes: 97 items in 2023, 85 in 2024 and 67 in 2025.

Is fear of public speaking rising?

Possibly, but three waves are not enough to call it a trend. The 2024-to-2025 change (+4.7 points, from 29.0% to 33.7%) is about the size of the surveys' sampling error: Chapman's methods reports give ±4.0 points for 2024 and ±3.6 for 2025, and a rough margin for the difference between two independent surveys is the square root of 4.0² + 3.6², which is about 5.4 points. That margin is stated for estimates near 50%, so it is generous for a figure near 30%. The steadier signal is the "very afraid" share, which moved from 10.4% to 11.9% to 15.5% in three successive years. Write "rose to 33.7% in 2025" rather than "fear of public speaking is surging".

Is public speaking the number one fear?

No. In the 2025 Chapman survey public speaking ranked 46th of 67 fears, below corrupt government officials (69.1%, first), a loved one becoming seriously ill (58.9%) and dozens of financial, environmental and security fears. The "number one fear, worse than death" line comes from a 1973 survey and is usually misquoted.

According to Karen Kangas Dwyer and Marlina Davidson, who set out to trace it, the claim began with a 1973 American fears study by R. H. Bruskin Associates that appeared in the London Sunday Times and was then repeated in communication textbooks as "people fear public speaking more than death" (Dwyer and Davidson, Communication Research Reports, 2012). The original 1973 report is not available online; this page relies on their published account of it. When they repeated the exercise with 815 college students, public speaking was ticked as a fear more often than any other item, death included, but when students had to name their top fear, they chose death most often.

So the accurate version is narrower than the slogan: on a checklist, about as many people tick public speaking as tick death, and sometimes more; asked what they fear most, people do not choose public speaking. Chapman's data show the same pattern. Chapman's own key findings note, "anticipating a frequently asked question", that public speaking ranked 46th in 2025.

Share of US adults afraid or very afraid, from Chapman's complete lists of fears for each year. The gaps in 2023 and 2024 are smaller than each survey's margin of error, so the defensible sentence is "about as many Americans fear public speaking as fear dying, and neither is near the top of the list".

Fear vs disorder: how common is social anxiety disorder?

About 7.1% of US adults had social anxiety disorder in the past year and 12.1% at some point in life, according to the US National Institute of Mental Health, based on the National Comorbidity Survey Replication (2001–2003). Across 28 community surveys in the World Mental Health Survey Initiative, the combined 12-month rate was 2.4% and the lifetime rate 4.0%.

A fear and a disorder are different measurements, and confusing them is a frequent mistake on this topic, along with the "number one fear" line. Chapman asks people how afraid they are; the National Comorbidity Survey Replication used structured diagnostic interviews. In that survey, 21.2% of US adults reported a fear of public speaking or performance at some point in life that was excessive and caused real distress or avoidance, the most common of 14 social fears measured; speaking up in a meeting or class came second at 19.5% (Ruscio et al., 2008). About half of those people (50.6%) met the criteria for social anxiety disorder over their lifetime, which equals 10.7% of all adults. Seen from the other side, 88.7% of people with lifetime social anxiety disorder had a public speaking or performance fear.

The global rate is lower than the US rate, and the authors of the cross-national study report that rates were lowest in low and lower-middle income countries and in the African and Eastern Mediterranean regions, and highest in high-income countries and in the Americas and Western Pacific regions (Stein et al., 2017). Two cautions apply to every clinical number in this section: the US interview data are more than 20 years old, and none of these surveys is a measure of how many people in your audience are nervous today.

Who is more affected: age and sex

Women and younger adults report social anxiety disorder more often: in the US past-year data, 8.0% of women and 6.1% of men had the disorder, falling from 9.1% at ages 18–29 to 3.1% at 60 and over (NIMH, NCS-R 2001–2003). Among US adolescents aged 13–18, lifetime prevalence was 9.1% (11.2% of girls, 7.0% of boys).

Source: NIMH, Social Anxiety Disorder statistics. Among adults with past-year social anxiety disorder, NIMH reports impairment as 29.9% serious, 38.8% moderate and 31.3% mild. Chapman does not publish the public speaking item by sex or age in its summary documents; the full datasets are archived and released on a rolling basis at the Association of Religion Data Archives, so a breakdown may become possible but is not reported here. Some statistics pages present "8% of women and 6% of men fear public speaking"; those are the NIMH social anxiety disorder figures, relabelled.

Attention span statistics: what the 47 seconds really measures

The "47-second attention span" comes from Gloria Mark, professor emerita of informatics at the University of California, Irvine, and it measures how long people stay on one computer screen before switching, not how long they can concentrate. In her studies, average time on a single screen fell from about two and a half minutes (150 seconds) in 2003 to 47 seconds across studies from 2016 to 2020.

Mark describes the measurement in her own words in a University of California interview: her first paper on the subject was published in 2004 from a study conducted in 2003 (González and Mark, CHI 2004), and later studies logged how long people stayed on one screen at work (UC Newsroom, 2 January 2024). She adds that the figure appears to have reached a steady state of about 47 seconds for roughly five years. That is a finding about multitasking at a computer, where switching is easy and often self-interrupted. It says nothing direct about how long an audience can follow a speaker in a room, where there is one screen and nothing to switch to except a phone.

For presenters the useful lesson is about competition, not capacity: when people have a device in front of them, every slide has to give them a reason not to switch. That argues for short slide titles that state the point, one idea per slide and a change of activity at planned moments, which our guide to structuring a presentation turns into frameworks.

The goldfish attention span myth, traced

The claim that human attention spans fell from 12 seconds in 2000 to 8 seconds, "less than a goldfish", has no research behind it. BBC Radio's More or Less traced it in 2017 to a 2015 Microsoft Canada consumer insights report, which did not produce the figure itself but cited a website called Statistic Brain; the sources Statistic Brain listed could find no record of research supporting it.

Simon Maybin's investigation for the BBC World Service programme found that the Microsoft Canada team had surveyed 2,000 Canadians and studied the brain activity of 112 people, but that the headline attention-span number was imported from Statistic Brain, whose listed sources (the US National Center for Biotechnology Information and the Associated Press) could not find the research (BBC News, 10 March 2017). Psychologists interviewed for the piece said an "average attention span" is not something they measure, because attention depends on the task; a fish-behaviour researcher added that goldfish are a standard model species in memory research. If you need an attention number for a talk, use none, or use Mark's screen-switching figure with its correct description.

The 7-38-55 rule: what Mehrabian actually found

The 7-38-55 rule says that 7% of a message's meaning comes from words, 38% from tone of voice and 55% from facial expression, but its author, psychologist Albert Mehrabian, states that it applies only to communications about feelings and attitudes, such as liking or disliking. It is not a rule about presentations, and it does not mean 93% of communication is non-verbal.

On his own website, Mehrabian gives the equation (total liking = 7% verbal + 38% vocal + 55% facial) and states that it and similar equations were derived from experiments on communicating feelings and attitudes; unless a speaker is talking about their feelings or attitudes, the equations do not apply (Albert Mehrabian, Silent Messages description). The experiments looked at inconsistent messages, where words and tone or face disagreed, and his book discusses them on pages 75 to 80. A presentation in which the content carried only 7% of the meaning would need no slides, no data and no argument; the rule is quoted in public speaking courses because it sounds precise, not because it measures speeches.

Do students stop listening after 10 minutes?

The research does not support a fixed 10–15-minute attention limit in lectures. Karen Wilson and James Korn reviewed the studies behind the claim, including note-taking studies, classroom observation, self-reports and physiological measures, and concluded that they provide little support for the belief that attention declines after 10 to 15 minutes (Wilson and Korn, Teaching of Psychology, 2007).

Wilson and Korn also found that most of the studies ignored individual differences between students, and they advised instructors to check whether students were capturing the relevant content in their notes rather than relying on a timer. The practical version for speakers is to plan changes of pace by content (a new question, a demonstration, a short exercise) and to check understanding, not to assume the room switches off at minute ten. Our guide to how many slides per minute covers pacing by talk length.

The myth ledger: six public speaking claims and their sources

Six claims about public speaking and attention are repeated far more often than they are checked. The ledger below gives each claim, where it came from, what the original source actually says and a verdict; every origin was opened on 14 October 2026.

One more number deserves a line of its own because it leads so many pages: "77% (or 73%, 74%, 75%) of people fear public speaking". We found no primary study behind it, and NIMH's social anxiety statistics page, sometimes listed among the sources of pages that use it, contains no public speaking percentage at all. This page does not use it. If you need a "most people" figure, the defensible one is Chapman's: 68.5% of US adults were at least slightly afraid of public speaking in 2025.

Which public speaking statistic to cite, sentence by sentence

Choose the statistic by the sentence you want to write, then copy the wording in the third column; most misquotes happen when a correct number is attached to the wrong claim. The table covers the eight sentences people most often want to write about public speaking.

What the numbers mean for an audience of 10, 30, 100 or 500

Applied to a room, the 2025 rates suggest that about one in three people would describe themselves as afraid of speaking in front of it, about two in three as at least slightly afraid, and about one in fourteen might meet the criteria for social anxiety disorder in a given year. These are population rates used as an illustration, not a prediction for any specific group.

Arithmetic: multiply the number of people by the rate, for example 30 × 0.337 = 10.1. The US rates come from different years (2025 for fear, 2001–2003 for the disorder), and any real group differs from the national population: a class of first-year students, a sales team and a conference of trainers will not match the average.

Three practical takeaways follow from the data. First, nervousness is normal, so plan for it: learn the first two sentences word for word, as our guide to starting a presentation recommends. Second, the attention research is about competition, so give every slide one point and a title that states it; the professional design rules show how. Third, people train for this at scale: Toastmasters International reports more than 265,000 members in more than 13,800 clubs in nearly 150 countries (fact sheet, revised December 2025), and rehearsal in front of real people remains the most available form of practice. When you close, end on the action rather than a statistic, as in our guide to ending a presentation.

How to put a public speaking statistic on a slide

A statistic on a slide needs four things besides the number: the population, the year, the exact measure and a source line. "33.7% of US adults were afraid or very afraid of public speaking in 2025" is a citable statement; "77% fear public speaking" is not, because it has no population, no year, no measure and no source.

A source line in the format "Source: Organization, title (year)" is the convention most data slides use; our guide to AI presentations with research and citations covers it in depth, and how to present data covers chart choice. In NOXI, slides with figures in a deck started from a prompt carry a source line linking to where each number came from; when you supply the numbers yourself, give the source line with them, as in the prompts below.

Key terms

How to cite this page

Cite the primary source for any single number, and cite this page when you use its compilation, the myth ledger, the citation table or the room calculations. Always include the "last checked" date, because the Chapman figures change every year.

Methodology and update log

Every figure on this page was read from its primary source on 14 October 2026: Chapman's published lists, key findings and SSRS methods reports for 2023, 2024 and 2025; the peer-reviewed papers by Ruscio et al. (2008), Stein et al. (2017), Dwyer and Davidson (2012), and Wilson and Korn (2007); NIMH's statistics and publications pages; the BBC investigation; Mehrabian's own page; and Gloria Mark's University of California interview.

• Percentages are reported as the source publishes them, to one decimal place. The 68.5% figure is our sum of three published shares (15.5% + 18.1% + 34.9%) from the 2025 weighted frequency table.
• Ranks for 2023 and 2025 are Chapman's; the 2024 rank is the position in Chapman's own highest-to-lowest list. Ranks are not compared across years because the set of fears changes.
• The 5.4-point margin for year-to-year change is our approximation from the two published margins of error, not a figure Chapman publishes.
• Excluded: any percentage we could not trace to a named study with a sample and year, including the 73–77% fear figures, and any count of "presentations given per day".
• Data age: the US clinical figures come from interviews conducted in 2001–2003; they remain the figures NIMH publishes.

Update log. 14 October 2026: first edition, with Chapman's 2025 wave (Wave 11) as the latest published on Chapman's survey page. Chapman says data are usually collected in the spring and released the following fall; the 2025 results were released on 21 October 2025 (Chapman Newsroom). When the 2026 wave appears, the trend table, the rank comparison, the figures and the quick answer will be updated on this URL and the change recorded here.

Three AI prompts for slides with statistics

The three prompts below give the model the numbers and the source lines, so it arranges evidence rather than inventing it; keep the brackets for your own audience and length.

How to build a data-backed slide set in NOXI

NOXI builds a full deck from a prompt or from your files and can match the style of a template you already use. It is free, with no card and no watermark; uploading files needs a free account, and a full deck uses about 25 of the 1,000 monthly credits.

Frequently asked questions

What percentage of people fear public speaking?

In the 2025 Chapman University Survey of American Fears, 33.7% of US adults said they were afraid or very afraid of public speaking, and 68.5% said they were at least slightly afraid. The survey asked 1,015 adults from a probability-based national panel between 24 March and 8 April 2025, with a margin of error of ±3.6 points. Figures of 73% to 77% that appear online have no traceable primary study.

Is public speaking the number one fear?

No. Public speaking ranked 46th of 67 fears in Chapman's 2025 survey; the top fear was corrupt government officials at 69.1%. The "number one fear" line goes back to a 1973 Bruskin survey in which people ticked public speaking more often than other fears on a list. When students in a 2012 replication study had to name a single top fear, they chose death most often.

Do people fear public speaking more than death?

On a checklist, about as many people name public speaking as name dying, and sometimes more: in Chapman's 2025 survey 33.7% feared public speaking and 28.6% feared dying, while in 2024 the order was reversed (29.0% and 31.6%). When students were asked to choose a top fear in a 2012 replication study, they chose death most often. Neither fear is near the top of the list.

How common is glossophobia?

Glossophobia is a popular label, not a diagnosis, so there is no official glossophobia rate. The closest measures are 33.7% of US adults afraid or very afraid of public speaking (Chapman, 2025), 21.2% with an excessive, distressing fear of public speaking or performing at some point in life, and 10.7% with social anxiety disorder involving that fear (National Comorbidity Survey Replication, 2001–2003).

How many people have social anxiety disorder?

About 7.1% of US adults had social anxiety disorder in the past year and 12.1% at some point in their lives, according to NIMH, using National Comorbidity Survey Replication data from 2001–2003. Across 28 community surveys in the World Mental Health Survey Initiative, the combined rates were 2.4% for the past 12 months and 4.0% for a lifetime. Rates are higher among women and younger adults.

Are women more afraid of public speaking than men?

For social anxiety disorder, yes: 8.0% of US women and 6.1% of men had it in the past year (NIMH, 2001–2003 data), and among adolescents the lifetime rates were 11.2% for girls and 7.0% for boys. Chapman does not publish its public speaking item by sex in its summary reports, so there is no current published sex breakdown for self-reported fear of public speaking.

What is the average attention span in 2026?

There is no scientific measure of an average human attention span; researchers say attention depends on the task. The often-quoted 47 seconds is Gloria Mark's measurement of how long people stay on one computer screen before switching, averaged across her studies from 2016 to 2020, down from about 150 seconds in 2003. It describes multitasking at a computer, not how long an audience can follow a talk.

Is the goldfish attention span real?

No. The claim that people have an 8-second attention span, shorter than a goldfish's 9 seconds, was traced by BBC Radio's More or Less in 2017 to a 2015 Microsoft Canada report that cited a website called Statistic Brain. The sources that website listed could find no research behind the figure, and fish researchers note that goldfish are a standard species for studying memory.

What does the 7-38-55 rule really mean?

It means that when words, tone of voice and facial expression disagree about a feeling or attitude, such as liking, listeners weigh words at 7%, voice at 38% and face at 55%. Albert Mehrabian, who derived it, states that the equations do not apply unless the speaker is communicating feelings or attitudes. It is not evidence that 93% of a presentation's meaning is non-verbal.

Sources & methodology

All sources were opened and checked on 14 October 2026. Ranks, the 68.5% sum, the room calculations and the 5.4-point margin are our calculations from the published figures, shown with their arithmetic above. Priya and her workshop are fictional.

• [1] 2025 results, public speaking ranked 46th at 33.7%, sample of 1,015 and margin of error ±3.6% — Chapman University, Survey of American Fears 2025: Key Findings.
• [2] Complete 2025 list of 67 fears with shares afraid or very afraid — Chapman University, The Complete List of Fears 2025 (Wave 11), highest to lowest.
• [3] 2025 fieldwork dates, panel design, question wording, design effect, margin of error and weighted answer split for public speaking and dying — SSRS for Chapman University, Survey of American Fears Methods Report 2025.
• [4] Public speaking 29 in 2024 and 33.7 in 2025, a 4.7-point difference — Chapman University, Survey of American Fears Wave 10 (2024) and Wave 11 (2025) Compared.
• [5] 2024 list of 85 fears, public speaking 29.0% and dying 31.6% — Chapman University, The Complete List of Fears 2024 (Wave 10).
• [6] 2024 sample of 1,008, fieldwork 21 March–1 April 2024, margin of error ±4.0 points and public speaking answer split — SSRS for Chapman University, Survey of American Fears Methods Report 2024.
• [7] 2023 list of 97 fears, public speaking 53rd at 28.7% and dying 26.2% — Chapman University, The Complete List of Fears 2023 (Wave 9).
• [8] 2023 sample of 1,014, fieldwork 27 March–4 April 2023, margin of error 3.7 points and public speaking answer split — SSRS for Chapman University, Survey of American Fears Wave 9 Methodology Report 2023.
• [9] Survey conducted annually since 2014, spring data collection and fall releases, data archived at the ARDA; latest wave listed 2025 — Chapman University, Survey of American Fears; 2025 release — Chapman Newsroom, What Americans Fear Most in 2025 (21 October 2025).
• [10] Lifetime public speaking or performance fear 21.2%, 50.6% of them with lifetime social phobia, 10.7% of all adults, 88.7% of people with social phobia; n = 9,282 — Ruscio AM, Brown TA, Chiu WT, Sareen J, Stein MB, Kessler RC, Social fears and social phobia in the USA: results from the National Comorbidity Survey Replication, Psychological Medicine 38(1), 2008, pp. 15–28.
• [11] Past-year and lifetime social anxiety disorder, by sex and age, impairment, adolescents — National Institute of Mental Health, Social Anxiety Disorder statistics.
• [12] Definition of social anxiety disorder and anxiety arising only in performance situations such as giving a speech — National Institute of Mental Health, Social Anxiety Disorder: More Than Just Shyness (revised 2025).
• [13] Cross-national 30-day, 12-month and lifetime prevalence of 1.3%, 2.4% and 4.0% across 28 surveys and 142,405 respondents — Stein DJ et al., The cross-national epidemiology of social anxiety disorder: data from the World Mental Health Survey Initiative, BMC Medicine 15:143, 2017.
• [14] Origin of "more feared than death" in the 1973 Bruskin study; replication with 815 students — Karen Kangas Dwyer and Marlina M. Davidson, Is Public Speaking Really More Feared Than Death?, Communication Research Reports 29(2), 2012, pp. 99–107.
• [15] The 8-second attention span and goldfish claim traced to Microsoft Canada and Statistic Brain — Simon Maybin, Busting the attention span myth, BBC News / More or Less (10 March 2017).
• [16] Time on a single screen: 150 seconds in the 2003 study, 47 seconds across 2016–2020 studies — University of California Newsroom, How to sharpen your attention and meet your goals in 2024 (2 January 2024).
• [17] The first study of screen switching — Victor M. González and Gloria Mark, "Constant, constant, multi-tasking craziness", Proceedings of CHI 2004, pp. 113–120.
• [18] The 7-38-55 equation and its limit to communications of feelings and attitudes — Albert Mehrabian, "Silent Messages": description and ordering information.
• [19] Little support for a decline in student attention after 10–15 minutes — Karen Wilson and James H. Korn, Attention during Lectures: Beyond Ten Minutes, Teaching of Psychology 34(2), 2007, pp. 85–89.
• [20] Toastmasters membership, clubs and countries — Toastmasters International Fact Sheet (Rev. 12/2025).

Turn these numbers into slides in NOXI, free: one statistic per slide, a title that states the finding and a source line under every figure, in your own template, exported as an editable PowerPoint or PDF.