owlfox.studioBig claims, checked.YouTube

Sources · Tested · Episode 2

Did We Stop Hanging Out? We Tested 22 Years of U.S. Data.

Every figure shown on screen in this episode, with the document, page and date it comes from, and a label for how strong the evidence is.

Americans age 15 and over spent less time socializing in 2025 than in 2003, mostly because fewer of them socialized on a given day. Those who did still spent almost two hours at it.

Published
29 September 2026
Length
9:04
Figures checked
26 September 2026
Documents
8
Watch the episodeJump to the documents

Verdict

  • HOLDS“Americans spent less time socializing in 2025 than in 2003”
  • PARTLY“Did Americans stop hanging out?”

On screen in this episode

  • DOCUMENTED34The document shows it, and you can check it yourself.
  • NOT SHOWN10Something that matters is missing from every source we checked.
  • OUR CALCULATION7Our own arithmetic from the sourced numbers. The method is below.
  • missing2A value is absent from the dataset. It is shown as missing, never as zero.

How we label evidence

NOT SHOWNWhat isn’t shown

  • 1:17FRIENDSHIP · LONELINESS · NOT MEASURED HERE
  • 4:39SOME STOPPED? · MANY WENT LESS OFTEN? · A MIX OF BOTH?
  • 5:43YOUNGER ADULTS? · A NATIONAL AVERAGE CANNOT SAY
  • 6:5722-YEAR NATIONAL DECLINE · HOW MUCH DID PHONES CAUSE? · NOT IDENTIFIED HERE
  • 7:12PLAUSIBLE IS NOT PROVEN
  • 7:21PHONES MAY MATTER · HOW MUCH? · THIS TEST CANNOT SAY
  • 7:362 / TIME DOES NOT MEASURE QUALITY
  • 7:473 / UNITED STATES · NOT A RESULT FOR EVERY COUNTRY
  • 7:53TIME USE · LONELINESS · NOT MEASURED HERE · NEEDS DIFFERENT EVIDENCE
  • 8:42THE CAUSE IS STILL OPEN · THE DIARIES CANNOT CONVICT

On screen, in order

  1. 0:00

    100 PEOPLE · 2003 · AN AVERAGE DAY · ≈40 SOCIALIZED THAT DAY

    DOCUMENTED

    Rounded illustration of 40.2%.

    • [1] BLS series, share who socialized, 2003: 40.2%
  2. 0:08

    100 PEOPLE · 2025 · AN AVERAGE DAY · ≈30 SOCIALIZED THAT DAY

    DOCUMENTED

    Rounded illustration of 30.4%.

    • [1] BLS series, share who socialized, 2025: 30.4%
  3. 0:14

    STILL ALMOST TWO HOURS · AMONG THOSE WHO SOCIALIZED

    DOCUMENTED

    116.4 and 115.2 minutes that day.

    • [2] BLS series, time among participants, 2003: 1.94 hours; 2025: 1.92 hours

    How we worked this out

  4. 0:47

    ONE FULL DAY · 04:00 → NEXT DAY · 04:00 · 24 HOURS · AGE 15+

    DOCUMENTED

    Each respondent reports the day before the interview, from 4 a.m. to 4 a.m.

    • [3] BLS ATUS technical note, 2025, About the questionnaire; About the sample
  5. 0:58

    SOCIALIZING · FACE TO FACE · VISITING PEOPLE · SOCIAL EVENTS

    DOCUMENTED
    • [4] BLS ATUS activity lexicon, Socializing and communicating
  6. 1:07

    COUNTED SEPARATELY

    DOCUMENTED

    Phone calls, mail and email are separate activity categories.

    • [4] BLS ATUS activity lexicon
  7. 1:11

    THE MAIN ACTIVITY · COOKING · TALKING CAN BE MISSED

    DOCUMENTED

    The estimates count only the main activity at each moment.

    • [3] BLS ATUS technical note, 2025, Definitions: primary activity
  8. 1:17

    FRIENDSHIP · LONELINESS · NOT MEASURED HERE

    NOT SHOWN

    The diary records time use, not friendships or loneliness.

    • Checked[3] BLS ATUS technical note, 2025
  9. 2:00

    2003–2025 (chart of average socializing minutes per person per day)

    DOCUMENTED
    • [5] BLS series, average time, annual estimates 2003–2025

    How we worked this out

  10. 2:05

    2020 MISSING · MISSING ≠ ZERO

    missing

    The annual series has no 2020 value. Data collection was suspended from mid-March to mid-May 2020.

    • Missing in[5] BLS series, average time, 2020: (M)
    • [6] BLS ATUS data files page
  11. 2:19

    SOCIALIZING · AVERAGE DAY · 2003 46.8 → 2025 34.8 MIN / PERSON / DAY

    DOCUMENTED
    • [5] BLS series, average time, 2003: 0.78 hours; 2025: 0.58 hours

    How we worked this out

  12. 2:29

    −12 MIN / DAY · ≈−26%

    OUR CALCULATION
    • [5] BLS series, average time

    How we worked this out

  13. 2:40

    OUR CALCULATION · 12 MIN PER DAY → ≈73 HOURS OVER A YEAR · 12 × 365 ÷ 60 · NOT A LOSS MEASURED FOR EACH PERSON

    OUR CALCULATION
    • [5] BLS series, average time

    How we worked this out

  14. 2:51

    THE DROP IS CLEAR · −12.0 MIN · −15.05 TO −8.95 · 0 = NO CHANGE

    OUR CALCULATION
    • [5] BLS series, average time, standard errors 2003 and 2025

    How we worked this out

  15. 2:59

    NOT A SMOOTH SLIDE · ALREADY LOWER IN 2019 · 34.8

    DOCUMENTED
    • [5] BLS series, average time, annual estimates 2003–2025

    How we worked this out

  16. 3:10

    LESS TOTAL SOCIALIZING TIME · 46.8 → 34.8 · HOLDS

    DOCUMENTED

    The verdict on the narrow claim.

    • [5] BLS series, average time, 2003: 0.78 hours; 2025: 0.58 hours

    How we worked this out

  17. 3:20

    SHARE WHO SOCIALIZED · 2003 40.2% → 2025 30.4% · ON AN AVERAGE DAY

    DOCUMENTED
    • [1] BLS series, share who socialized, 2003: 40.2%; 2025: 30.4%
  18. 3:30

    −9.8 PERCENTAGE POINTS

    OUR CALCULATION
    • [1] BLS series, share who socialized

    How we worked this out

  19. 3:34

    100 PEOPLE · ≈10 FEWER IN 100

    OUR CALCULATION

    Rounded illustration of 9.8 percentage points; each year surveys different people.

    • [1] BLS series, share who socialized

    How we worked this out

  20. 3:41

    AMONG PEOPLE WHO SOCIALIZED · 2003 116.4 MIN → 2025 115.2 MIN

    DOCUMENTED
    • [2] BLS series, time among participants, 2003: 1.94 hours; 2025: 1.92 hours

    How we worked this out

  21. 3:52

    CHANGE IN MINUTES · −1.2 MIN · −8.32 TO +5.92 · NO CLEAR CHANGE

    OUR CALCULATION
    • [2] BLS series, time among participants, standard errors 2003 and 2025

    How we worked this out

  22. 3:58

    ALMOST TWO HOURS · BOTH YEARS

    DOCUMENTED
    • [2] BLS series, time among participants, 2003: 1.94 hours; 2025: 1.92 hours

    How we worked this out

  23. 4:02

    HOW MANY? · FEWER ON A GIVEN DAY

    DOCUMENTED

    Daily participation, not a count of people who never socialize.

    • [1] BLS series, share who socialized, 2003: 40.2%; 2025: 30.4%
  24. 4:08

    DIFFERENT SAMPLES · 2003 · 2025 · NOT FOLLOWED YEAR TO YEAR

    DOCUMENTED

    Each respondent is interviewed once, about one day.

    • [3] BLS ATUS technical note, 2025, About the sample
  25. 4:39

    SOME STOPPED? · MANY WENT LESS OFTEN? · A MIX OF BOTH?

    NOT SHOWN

    These data do not identify the proportions.

    • Checked[1] BLS series, share who socialized
  26. 4:51

    AFTER THE LOW POINT · 2022 33.6 → 2025 34.8 MIN / PERSON / DAY · LOWEST PUBLISHED YEAR

    DOCUMENTED
    • [5] BLS series, average time, 2022: 0.56 hours; 2025: 0.58 hours

    How we worked this out

  27. 5:05

    2022 28.7% → 2025 30.4% · SOCIALIZED THAT DAY

    DOCUMENTED
    • [1] BLS series, share who socialized, 2022: 28.7%; 2025: 30.4%
  28. 5:13

    DECLINE ALREADY UNDERWAY · 2003 46.8 → 2019 38.4 MIN / PERSON / DAY

    DOCUMENTED
    • [5] BLS series, average time, 2003: 0.78 hours; 2019: 0.64 hours

    How we worked this out

  29. 5:19

    2003 40.2% → 2019 34.8% · SOCIALIZED THAT DAY

    DOCUMENTED
    • [1] BLS series, share who socialized, 2003: 40.2%; 2019: 34.8%
  30. 5:24

    2025 STILL BELOW 2019 · TIME AND PARTICIPATION

    DOCUMENTED
    • [5] BLS series, average time, 2019: 0.64 hours; 2025: 0.58 hours
    • [1] BLS series, share who socialized, 2019: 34.8%; 2025: 30.4%
  31. 5:29

    THE DECLINE STARTED EARLIER · PANDEMIC

    DOCUMENTED

    The timing does not identify the cause or its size.

    • [5] BLS series, average time, annual estimates 2003–2019
  32. 5:43

    YOUNGER ADULTS? · A NATIONAL AVERAGE CANNOT SAY

    NOT SHOWN

    The national average does not show which age group changed most.

    • Checked[5] BLS series, average time
  33. 5:49

    A SEPARATE STUDY · 23–38 · SAME AGE RANGE · 2003 / 2019

    DOCUMENTED
    • [7] BLS Monthly Labor Review, 2022
  34. 6:05

    GENERATION X 2003 · MILLENNIALS 2019 · AGE 23–38 · −8 MIN / DAY

    DOCUMENTED
    • [7] BLS Monthly Labor Review, 2022, socializing and communicating: 8 fewer minutes a day
  35. 6:09

    DIFFERENCE CAME FROM WOMEN

    DOCUMENTED
    • [7] BLS Monthly Labor Review, 2022, millennial women 34 minutes, Generation X women 50 minutes
  36. 6:33

    RANDOM ASSIGNMENT · FACEBOOK ON · PAID TO DEACTIVATE · 4 WEEKS

    DOCUMENTED
    • [8] Allcott et al., American Economic Review 2020
  37. 6:41

    MORE OFFLINE TIME · FAMILY & FRIENDS · TELEVISION ALONE · OFFLINE ≠ ALWAYS SOCIAL

    DOCUMENTED

    Measured during the four-week experiment.

    • [8] Allcott et al., American Economic Review 2020, abstract
  38. 6:53

    EVIDENCE FOR SOME PEOPLE · 4-WEEK EXPERIMENT

    DOCUMENTED
    • [8] Allcott et al., American Economic Review 2020
  39. 6:57

    22-YEAR NATIONAL DECLINE · HOW MUCH DID PHONES CAUSE? · NOT IDENTIFIED HERE

    NOT SHOWN

    The experiment estimates its own effect, not a share of the national trend.

    • Checked[8] Allcott et al., American Economic Review 2020
    • [5] BLS series, average time
  40. 7:05

    RECORDING ≠ ASSIGNING · 04:00 TO 04:00

    DOCUMENTED

    The survey records what people did; nothing is assigned at random.

    • [3] BLS ATUS technical note, 2025, About the questionnaire
  41. 7:12

    PLAUSIBLE IS NOT PROVEN

    NOT SHOWN

    This comparison does not measure how much any possible cause contributed.

  42. 7:21

    PHONES MAY MATTER · HOW MUCH? · THIS TEST CANNOT SAY

    NOT SHOWN

    No quantified causal attribution.

    • Checked[8] Allcott et al., American Economic Review 2020
    • [5] BLS series, average time
  43. 7:28

    1 / THE MAIN ACTIVITY · DINNER · EVEN WITH CONVERSATION

    DOCUMENTED

    Talk over dinner is not counted as socializing when eating is the main activity.

    • [3] BLS ATUS technical note, 2025, Definitions: primary activity
  44. 7:36

    2 / TIME DOES NOT MEASURE QUALITY

    NOT SHOWN

    One diary day does not measure relationships.

  45. 7:47

    3 / UNITED STATES · NOT A RESULT FOR EVERY COUNTRY

    NOT SHOWN

    The survey covers people age 15 and over in the United States.

    • Checked[3] BLS ATUS technical note, 2025, About the sample
  46. 7:53

    TIME USE · LONELINESS · NOT MEASURED HERE · NEEDS DIFFERENT EVIDENCE

    NOT SHOWN

    A decline in this activity category is not a loneliness measure.

    • Checked[3] BLS ATUS technical note, 2025
  47. 8:00

    THE SAME KIND OF DIARY · 2020: MISSING

    missing

    Repeated annual estimates for the same activity category.

    • Missing in[5] BLS series, average time, 2020: (M)
    • [6] BLS ATUS data files page
  48. 8:13

    ≈−26% · AVERAGE SOCIALIZING TIME · 2003 → 2025

    OUR CALCULATION
    • [5] BLS series, average time

    How we worked this out

  49. 8:13

    DECLINE BEGAN BEFORE THE PANDEMIC

    DOCUMENTED
    • [5] BLS series, average time, 2003: 0.78 hours; 2019: 0.64 hours
    • [1] BLS series, share who socialized, 2003: 40.2%; 2019: 34.8%
  50. 8:25

    “STOPPED” GOES TOO FAR · ABOUT 3 IN 10 · ON A GIVEN DAY

    DOCUMENTED
    • [1] BLS series, share who socialized, 2025: 30.4%
  51. 8:31

    ALMOST TWO HOURS · NO CLEAR CHANGE SINCE 2003

    DOCUMENTED

    2025 participants averaged 115.2 minutes that day.

    • [2] BLS series, time among participants, 2025: 1.92 hours

    How we worked this out

  52. 8:38

    FEWER PEOPLE · ON A GIVEN DAY

    DOCUMENTED

    Participation is the larger observed change.

    • [1] BLS series, share who socialized
    • [2] BLS series, time among participants
  53. 8:42

    THE CAUSE IS STILL OPEN · THE DIARIES CANNOT CONVICT

    NOT SHOWN

    Possible contributors; no causal shares identified here.

Our calculations

  1. OUR CALCULATIONMinutes from hours

    BLS publishes these estimates in hours a day; minutes are hours × 60. Average time: 0.78 h = 46.8 min (2003), 0.64 h = 38.4 min (2019), 0.56 h = 33.6 min (2022), 0.58 h = 34.8 min (2025). Among participants: 1.94 h = 116.4 min (2003), 1.92 h = 115.2 min (2025).

    • [5] BLS series, average time
    • [2] BLS series, time among participants

  2. OUR CALCULATIONDifferences between years

    34.8 − 46.8 = −12.0 minutes, and −12.0 ÷ 46.8 = −25.6%, shown as ≈−26%. 30.4% − 40.2% = −9.8 percentage points, shown as about 10 fewer in 100 in a rounded illustration; the two years survey different people. Among participants, 115.2 − 116.4 = −1.2 minutes.

    • [5] BLS series, average time
    • [1] BLS series, share who socialized
    • [2] BLS series, time among participants

  3. OUR CALCULATION73 hours a year

    12 minutes a day × 365 days ÷ 60 = 73 hours. This scales the population average to a year; it is not a loss measured for each person.

    • [5] BLS series, average time

  4. OUR CALCULATIONSurvey uncertainty

    Approximate 95% intervals for the 2003–2025 change, from the standard errors BLS publishes with each estimate, treating the two annual samples as independent: change ± 1.96 × √(SE 2003² + SE 2025²). Average time: −12.0 minutes (−15.05 to −8.95), clearly below zero. Time among participants: −1.2 minutes (−8.32 to +5.92), which includes zero, so no clear change.

    • [5] BLS series, average time, SE 0.012 h (2003), 0.023 h (2025)
    • [2] BLS series, time among participants, SE 0.023 h (2003), 0.056 h (2025)

Documents

  • All figures are for the U.S. population age 15 and over, all days of the week.
  • In the video, only the verdicts and our own calculation carry a label on screen. This page gives every figure its evidence label.
  1. [1]

    U.S. Bureau of Labor Statistics, American Time Use Survey

    Percent participating on an average day: socializing and communicating (series TUU30105AA01013951)

    Published 25 June 2026 · Retrieved 26 September 2026

    2025 estimate released on 25 June 2026. The page lists every year with its standard error; 2020 is marked (M).

    All days, population age 15 and over.

    Used at 0:00 · 0:08 · 3:20 · 3:30 · 3:34 · 4:02 · 4:39 · 5:05 · 5:19 · 5:24 · 8:13 · 8:25 · 8:38

  2. [2]

    U.S. Bureau of Labor Statistics, American Time Use Survey

    Average hours per day for participants: socializing and communicating (series TUU20101AA01013951)

    Published 25 June 2026 · Retrieved 26 September 2026

    2025 estimate released on 25 June 2026. The page lists every year with its standard error; 2020 is marked (M).

    Total time that day among people who socialized on their diary day. All days, population age 15 and over.

    Used at 0:14 · 3:41 · 3:52 · 3:58 · 8:31 · 8:38

  3. [3]

    U.S. Bureau of Labor Statistics

    American Time Use Survey — 2025 Results: Technical Note (USDL-26-1022)

    Published 25 June 2026 · Retrieved 26 September 2026

    Used at 0:47 · 1:11 · 1:17 · 4:08 · 7:05 · 7:28 · 7:47 · 7:53

  4. [4]

    U.S. Bureau of Labor Statistics

    American Time Use Survey Activity Lexicon 2003–2025

    Retrieved 22 September 2026

    Used at 0:58 · 1:07

  5. [5]

    U.S. Bureau of Labor Statistics, American Time Use Survey

    Average hours per day: socializing and communicating (series TUU10101AA01013951)

    Published 25 June 2026 · Retrieved 26 September 2026

    2025 estimate released on 25 June 2026. The page lists every year with its standard error; 2020 is marked (M).

    All days, population age 15 and over. BLS publishes hours; the video shows minutes (hours × 60).

    Used at 2:00 · 2:05 · 2:19 · 2:29 · 2:40 · 2:51 · 2:59 · 3:10 · 4:51 · 5:13 · 5:24 · 5:29 · 5:43 · 6:57 · 7:21 · 8:00 · 8:13

  6. [6]

    U.S. Bureau of Labor Statistics

    ATUS 2003–2025 Multi-Year Microdata Files

    Retrieved 26 September 2026

    Says data collection was suspended in 2020 from mid-March to mid-May.

    Used at 2:05 · 8:00

  7. [7]

    U.S. Bureau of Labor Statistics, Monthly Labor Review

    Time use of millennials and Generation X: differences across time (Michelle Freeman · doi:10.21916/mlr.2022.3)

    Published January 2022 · Retrieved 26 September 2026

    Used at 5:49 · 6:05 · 6:09

  8. [8]

    American Economic Review

    The Welfare Effects of Social Media (Allcott, Braghieri, Eichmeyer and Gentzkow · vol. 110, no. 3)

    Published March 2020 · Retrieved 26 September 2026

    Randomized experiment: Facebook users deactivated their accounts for the four weeks before the 2018 U.S. midterm election.

    Used at 6:33 · 6:41 · 6:53 · 6:57 · 7:21

Corrections

No corrections so far.

Found an error? Email [email protected] with the time stamp and your source, or leave a comment under the video.