About the evidence
Updates & corrections
Changes should be as traceable as the statistics themselves. Release dates below use UTC; data periods remain attached to each finding.
2 October 2026 UTC · Existing guide example versioned
The Reddit company figure in the user-count guide now has a stored source, metric, series and observation in company editorial release company-editorial-2026-10-02. Its value, Q2 2026 period and original disclosure are unchanged. The new record preserves the September 30 access and October 1 source recheck separately from this integration review; it does not add a live feed or downloadable company dataset. Existing Pinterest and Snapchat records are unchanged.
Clarified adult-survey scope: ages 18–19 are included in the 18+ population, while people under 18 are excluded. The adult age bands do not estimate teenagers as a separate group. No source estimates changed.
October 2026 · Teen frequency and reported time added
Added CDC high-school frequency with the 2025 hourly-use estimate, published confidence intervals and matching 2023 comparison. CDC’s test found no statistically significant change. The older several-times-daily measure remains separate, preventing a false decline caused by switching thresholds.
Added a limited Gallup time-spent citation for U.S. teenagers aged 13–19 in summer 2023, with source-reported uncertainty and the selected-platform limitation. No Gallup dataset or chart export is enabled. Source checks and rights decisions remain distinct from independent human review.
1 October 2026 · Adult visit frequency added
Connected NCI HINTS 7 adult visit frequency: five national source-codebook percentages from 2024, with standard errors calculated by SocialUsage using the supplied replicate weights. Missing item responses remain in the denominator. The release includes versioned aggregate CSV downloads and source lineage, with no respondent-level distribution. The questionnaire-reference correction published in v2 is recorded below.
1 October 2026 · U.S. coverage becomes the starting point
Added the NTIA U.S. dataset and made it the default explorer. Its 75 published aggregate observations cover national, demographic and state estimates with source standard errors. The reference period remains November 2023, with internet users aged 15+ as the denominator and a six-month activity window. Versioned CSV and original SVG exports retain attribution and scope.
Added limited Facebook and TikTok adult adoption briefs, bringing the Pew selection to five briefs and eleven source observations. No new age or historical cells were imported for those platforms. The U.S. age-gap analysis reuses the existing Instagram, YouTube and Reddit age values.
The European explorer remains available as separate coverage. U.S. and European values keep their distinct populations and activity windows. Source decisions and reuse limits.
1 October 2026 · Company platform briefs added
Added Pinterest and Snapchat briefs from original Q1 and Q2 2026 disclosures. Each provides a direct answer, a compatible two-quarter table and the latest regional figures, with definitions, sources and reuse limits. Pinterest’s monthly measure remains separate from Snapchat’s daily measure.
These are editorial source checks, not automated company API imports. Company-source downloads and image exports are disabled. The reading layout now keeps the headline answer and citation controls visible, with additional methods and tables available in expandable sections.
1 October 2026 · Initial research edition
Introduced the selected Eurostat 2023–2025 participation dataset and three U.S. platform briefs based on Pew’s 2025 survey. The Eurostat cohort contains 588 records, including seven confidential France 2024 cells.
Known source flags are preserved: France 2025 has a break in series; selected Ireland age groups in 2023–2024 are low reliability. The initial edition includes an EU age-gap calculation with explicit input records.
Verification was performed with an automated assistant. No independent human editorial approval is recorded. The site is prepared for review before a production launch.
Dataset release and lineage · Editorial and correction process
Corrections
2 October 2026 UTC: HINTS release hints7-2024-v2 corrects the questionnaire’s printed-page reference from 3 to 4. PDF page 5 and item B12a were already correct; values, uncertainty and denominator are unchanged. The original v1 release remains citable. Source revisions are logged separately from errors in SocialUsage calculations or presentation.
Freshness
NTIA’s published selection describes November 2023; its availability does not establish a released 2025 dataset. HINTS 7 describes March–September 2024, despite the package update in August 2025. CDC’s hourly-use series contains 2023 and 2025; Gallup’s time estimate remains summer 2023. The Eurostat cohort and Pew briefs end in 2025; Pinterest and Snapchat cover company disclosures through Q2 2026. A new evidence-check date does not update the data. Refreshes follow actual source releases.