Our sources

We only publish numbers we can trace to a primary source: the original study, official statistics, or the organization's own reporting. Secondary retellings, influencer polls, and unattributed "studies show" claims are rejected at the fact-gate.

Primary sources currently cited

• National Center for Family & Marriage Research (NCFMR), Bowling Green State University, Family Profile FP-25-03 (2023)U.S. Census Bureau American Community Survey (ACS) 1-year estimates, 2023, accessed via IPUMS-USA; rate = remarriages per 1,000 previously married persons
• National Center for Health Statistics (Bramlett & Mosher, Vital and Health Statistics, Series 23, No. 22) (2002)Life-table estimates from the 1995 National Survey of Family Growth, Cycle 5 — face-to-face interviews with 10,847 women aged 15-44 in the United States. The report defines 'disruption' as separation OR divorce, and measures the probability of disruption within a fixed number of years, not over a lifetime.

What we rejected, and why

This list matters as much as the one above. These are figures and claims that circulate widely in writing about remarriage and second marriages, and that we could not trace to a primary source, so they do not appear anywhere on this site.

• Second-wedding vows should run "roughly a minute each", per what officiants describe as workable. — Duration claim with no named study, survey or published methodology. Not traceable to a primary or authoritative source. Fix: drop the figure and keep the qualitative form ("shorter than most people draft"), or cite a named officiant-association guide with a year and URL.
• Things that vary between remarriages "appear repeatedly in the research literature" — money committed elsewhere, children, housing, a former spouse in the logistics. — The "studies show" pattern: no study, author, year or URL. Same class of claim the page itself declares it will not print. Fix: rephrase to what the site can stand behind — "things couples in this situation routinely have to work around" — or name one study with a year and URL.
• Published patterns link earlier marriage ages with higher rates of ending. — Associational claim about a predictor, attributed to no named body. Not covered by the content-spec §0 allowance, which permits only a qualitative direction for remarriage vs first marriage. Fix: cite NCHS/NSFG or Census with year and URL, or delete the sentence and keep "age on its own is a weak reason to move a wedding date".
• Emotionally, families overwhelmingly say yes — stepchildren remain stepchildren after a spouse dies. — "Overwhelmingly" asserts a distribution, in an answer written to be extracted by AI overviews. Fix (no source needed): "Many families say yes."
• Ambiguity over whether "my children" includes stepchildren is one of the more common sources of contested estates in blended families. — Frequency ranking about litigation with no named source. Fix (no source needed): "a well-known source of dispute".
• "Men remarry somewhat sooner on average than women" — presented as the pattern "most sources agree on". — Timing claim asserted as cross-source consensus with no named source. The verified NCFMR FP-25-03 (2023) figure is a remarriage RATE, not an interval, so it cannot be used to prop it up. Fix: drop the timing sentence, or use the sourced rate — "NCFMR (Bowling Green State University), on 2023 ACS data, put the remarriage rate at 34.4 per 1,000 previously married men and 18.5 per 1,000 previously married women" — while stating that no reliable average interval exists.
• "Cohabiting without marrying is common in this group" (men who never remarry). — Population frequency claim with no source. Fix: drop it, or cite an NCFMR/Pew cohabitation profile by name, year and URL.
• "Registrars and family lawyers describe it as a small but steady occurrence" (remarrying the same person). — Frequency claim attributed to an unnamed professional body — the "studies show" pattern. Fix: keep only the verifiable part (marriage records generally do not record whether the new spouse is the same spouse, so the frequency is not measured) and delete the frequency assertion, including the bare "It is uncommon".
• "Children tolerate 'different parents, different rules' more easily than adults expect." — Behavioural claim about children asserted as fact, unsourced, and edging into reading other people's children (content-spec §0, ДІТИ). Fix: rephrase as reported practice — "households commonly describe leaving the two rule sets different rather than forcing a merge" — with no claim about how children experience it.
• "The single most common source of conflict families describe is not a child's behaviour but two adults discovering mid-argument that they never agreed who was allowed to respond to it." — Superlative frequency claim ("single most common") with no survey behind it. Fix: "A conflict families name often is …" — drop the ranking.
• "A promise said in front of guests creates no rights and no obligations towards a stepchild." — Categorical statement of legal effect without the jurisdiction hedge §0 requires. Fix: "A ceremony is not a legal act; a stepparent's standing comes from law that differs by jurisdiction — that question goes to a family lawyer."
• "About 60% (or 67%) of second marriages end in divorce, and about 73% of third marriages" — routinely attributed to the NCHS / Bramlett & Mosher report. — Checked against the PDF itself: no 60% figure, no 67% figure and no third-marriage estimate of any kind appears in that report. What it publishes are life-table probabilities that a second marriage has DISRUPTED (separated or divorced) within 3, 5 or 10 years, among women aged 15-44 interviewed in 1995. The folklore version fails four ways at once: it converts a fixed-horizon probability into a lifetime one, converts 'disruption including separation' into 'divorce', drops the women-only, ages-15-44, 1995 population, and invents a third-marriage number that was never estimated. The figures may be cited to REFUTE the folklore, never to support it. Fix: quote the approved 3-year and 5-year disruption entries with their population, or keep the qualitative direction only.

The rules

1. Named source in the same sentence as the number, with the year.
2. Primary source only — we link the origin, not an article about it.
3. Dated verification: every statistic carries its verification date on the statistics page.
4. Unverifiable → unpublished. No exceptions.